Faculty of Media Engineering and Technology

Acoustic Environment Classification

Classification of acoustic environment relies on choosing the right set of features from the captured audio stream to represent the ambient environment. Normally the audio signal is captured as a discrete-time waveform, which cannot be used in its raw form. It is rather transformed into another domain where their instantaneous time values are replaced by meaningful feature vectors. Some of such features are: Fast Fourier Transform (FFT) Coefficients, Cepestral Coefficients and Linear Predictive Coding (LPC) Coefficients. Not all features are of the same complexity nor providing the same performance.

Related Bachelor Projects:
- Comparing Different Features for Acc. Env. Class.
  The student is expected to carry out a comparative study among a number of the most commonly used features in acoustic classification systems and provide experimental results with supporting analysis.
  Project Tasks:
  1. Comparing Acoustic Environment Classification Techniques Research
    
    Supervised by:
    
    Dr. Hisham Othman
    
    Context-aware computing has become a reality by which wearable and mobile devices sense the surrounding conditions and change their behavior accordingly in order to provide the best experience to the user. Context can be environmental like lighting, noise, traffic condition and social situation, or computational like wireless technology. For example, a smart phone may vibrate or ring depending on whether the user is in a quiet or noisy environment. There exist several techniques to classify a given environment based on a predetermined set of environments, including Hidden Markov Model (HMM), Neural Nets, and k-Nearest Neighbor (KNN). The student is expected to conduct a study comparing classification techniques in terms of accuracy and complexity.
  2. Adaptive Acoustic Environment Classification Research
    
    Supervised by:
    
    Dr. Hisham Othman
    
    Acoustic environment classification systems face a major challenge of non-stationarity of the environments. Adapting to changes in the environment features is a not a trivial task, yet indispensable for maintaining the system accuracy. Adaptation involves readjusting system size and/or modifying the values of its parameters. The student is expected to investigate the latest adaptation techniques used in acoustic environment classification and discuss their abilities and limitations based on results obtained from his/her own simulation. The work can be carried out using Matlab or Equivalent scientific package.

Ambient Computing

An Ambient system communicates information in an indirect way. For example, a corridor's light could be linked to the weather outside the building. The sunnier it is, the brighter the light in the corridor, and the reverse.

Related Bachelor Projects:
- Air Currents
  This project aims to define the parameters for using air currents to communicate information.
  Project Tasks:
  1. Using Mini-fans as an Ambient Display Research
    
    Supervised by:
    
    Dr. Wael Abouelsaadat
    
    You will build an app to control the operation of a set of mini-fans placed around the screen. The fans will be connected to a PC via a microcontroller board. Next, an HCI formal experiment will be designed and run to define the parameters for using air currents to communicate information. For example how close two fans can be placed and are still detectable as separate air sources. This project will give you experience in working with analog hardware and micro-controller interfacing boards. You will also learn how to conduct formal HCI experiments, perform statistical analysis on the data, and report the data in an academic format.

Automatic Configuration of Combinational Logic Circuits

Conventional logic design techniques tend to depend on domain-specific knowledge, which is somewhat limited both by the training and experience of the designer. While iterative heuristics, with little domain knowledge, allow the search for solutions in a much larger, and often richer, design space beyond the realms of conventional design techniques.

Related Bachelor Projects:
- Digital Circuit Design via Natural Computing Algorithms
  The project explores how biologically inspired algorithms such as swarm intelligence and evolutionary algorithms can be employed to tackle the digital design problem.
  Project Tasks:
  1. Digital Circuit Design using Evolutionary Algorithms for Research
    
    Supervised by:
    
    Prof. Dr. Hazem Abbas
    
    Genetic Algorithms employ the operators of natural evolution of population members through the concepts of selection, crossover, and mutation to find better solutions for digital circuit design. the process of evolutionary design is guaranteed to produce a functionally correct solution. Different flavors of evolutionary algorithms will be implemented in this task such as genetic algorithms, evolutionary strategies and memetic algorithms.
    Pre-requisite courses:
    
    CSEN301: Data Structures and Algorithms
    
    ELCT201: Digital Logic Design
    
    CSEN605: Digital System Design
  2. Ant Colony Optimization for optimal Digital Design Research
    
    Supervised by:
    
    Prof. Dr. Hazem Abbas
    
    Ant Colony Optimization (ACO) is a new meta-heuristic that combines distributed computation, positive feedback and constructive greedy heuristic in finding optimal solutions for combinatorial optimization problems. ACO involves cooperating agents: ants. The problem of logic design is needed to be formulated in such a way that it can be addressed by a set of cooperating ants to find an optimal circuit architecture.
    Pre-requisite courses:
    
    CSEN301: Data Structures and Algorithms
    
    ELCT201: Digital Logic Design
    
    CSEN605: Digital System Design

Combinatorial Algorithms

According to Kreher and Stinson (1999), combinatorial algorithms are classified into three classes: generation, enumeration, and search algorithms. Generation algorithms "[construct] all the combinatorial structures of a partiular type" (Kreher and Stinson, 1999, p. 1). These include algorithms for generating all subsets, partitions, or all permutations of a set. Enumeration algorithms "[compute the number of different structures of a particular type]" (Kreher and Stinson, 1999, p. 1). For example, finding the number of subsets of size k of a set of size n is an enumeration problem. Search algorithms "[find] at least one example of a structure of a particular type (if it exists)" (Kreher and Stinson, 1999, p. 1). These include, for example, algorithms for finding a clique of a certain size in a given graph. Many interesting combinatorial problems are, unfortunately, NP-hard.

Related Bachelor Projects:
- Constrained String Enumeration
  The enumeration problem for strings is one for counting the number of strings satisfying certain propeties. The simplest, and most common, such problem is the following. Given a sequence A1, A2, . . . , An of alphabets, find the number of strings of length n where symbol i comes from Ai (1 <= i < = n). This problem is solved directly by using the rule of products: |A1||A2| . . .|An|. The constrained string enumeration problem is similar to the above simple counting problem but with constraints of the following form: The number of discontinuous substrings with indices in a given set I are constrined to be at most mI.
  Project Tasks:
  1. Exploring Algorithms for Constrained String Enumeration Research
    
    Supervised by:
    
    Dr. Haythem Ismail
    
    One student will explore the, probably hard, problem of constrained string enumeration. Apparently, there is no mention of this problem in the literature. Thus, the work will involve designing, implementing, and evaluating exact and approximation algorithms for solving the problem.

Computer Supported Collaborative Work

Computer Supported Collaborative Work (CSCW) focuses on the development and study of systems designed to support collaboration between individuals or groups. Objective and subjective studies are often carried to evaluate, the psychological, social, and organizational impact of introducing a new system to ensure its’ effectiveness and efficiency.

Related Bachelor Projects:
- Collaborative Editing
  This topic focuses on aspects of collaborative editing
  Project Tasks:
  1. Multi-author Novels Research
    
    Supervised by:
    
    Dr. Wael Abouelsaadat
    
    A novel is typically written by one author and less often by a handful of authors. In this project, you are going to enable a new way of authoring novels, where users start adding chapters to an open content novel. Any user can start a novel by writing at least one chapter. Any other user can add a following chapter. A novel never ends and can have as many branches out of it as users like. Users vote on the best branch of chapters. This project will give you experience in working with various technologies involved in building a website. You will also be learning about visualization techniques for presenting the flow of a novel. Using CMS is not allowed.
- Video Conferencing
  The topic focuses on th aspects of Video Conferencing.
  Project Tasks:
  1. Connected Rooms Research
    
    Supervised by:
    
    Dr. Wael Abouelsaadat
    
    Video conferencing has helped families to connect with expatriate members. The goal of this project is to determine if connecting corresponding rooms in two houses belonging to the same family with an expatriate member will enhance the social interaction. To perform the study, a system will connect each two corresponding rooms; living room to living room, bedroom to bedroom, kitchen to kitchen, and dining room to dining room. There are two phases, a setup phase and a study phase. In the setup phase (1 month), the video conferencing system is installed and tested. In the study phase, a set of questionnaire and interviews will be regularly conducted for 3 months with the recruited family. To be able to conduct the interviews, you will need to quickly pick the HCI literature related to e-presence, collaborative work and video conferencing. The goal is to learn how to develop correct questions for the questionnaire and interviews and to identify what to ask for. Next, you will perform analysis on the data collected. This will be an opportunity for you to learn how to conduct long-term subjective studies, prepare questionnaires, conduct HCI interviews, and report your finding in an academic format.
- Email Communication
  The topic focuses on the aspects of Email Communication.
  Project Tasks:
  1. Your Handwriting Your Font Research
    
    Supervised by:
    
    Dr. Wael Abouelsaadat
    
    Technology is often criticized as being impersonal and cold. In this project, you are going to humanize the keyboard output! Instead of the system supplied designer fonts, you are going enable the user to use handwriting. The goal is to develop an application which automatically creates a font from scanned images of user hand writing. The application will also install the font on the user’s computer. In addition, you will conduct an HCI subjective study comparing designer fonts to hand writing based fonts, from the perspective of both the sender and recipient of emails. This project will train you in desktop application development. In addition, you will be learning about optical character recognition algorithms, as well as subjective HCI evaluations.

Constraint Reasoning

Constraint reasoning is concerned with the modelling and solving of combinatorial search and optimization problems in various applications areas such as scheduling, timetabling, placement, combinatorial designs, network design and portfolio optimization to name a few. It is also concerned with the design and implementation of programming languages that support the facilities to model and solve such problems.

Related Bachelor Projects:
- Simplex method
  The Simplex method is a global optimization method for linear systems of inequqalities. This project investigates the different pivot rules of the Simplex algorithm and their impact on the local search moves
  Project Tasks:
  1. Matrices in 2D and geometric reading Application
    
    Supervised by:
    
    Prof. Dr. Carmen Gervet
    
    The student will derive the first step towards a tool that shows how a set of 2D equations (over 2 variables X and Y) has an equivalent geometric reading in a 2D space.
  2. Comparing pivot rules of the Simplex algorithm Research
    
    Supervised by:
    
    Prof. Dr. Slim Abdennadher
    
    Prof. Dr. Carmen Gervet
    
    This project investigates the different pivot rules of the Simplex method for linear programming. The aim is to understand how they impact on the search toward the optimal solution. Ultimately we would also seek new pivot rules that would apply to symmetric combinatorial design problems and relate the LP model to a set based CSP model.
- Timetabling
  Project Tasks:
  1. Staff Scheduling System Research
    
    Supervised by:
    
    Prof. Dr. Slim Abdennadher
    
    Scheduling and timetabling is a complex problem that involves different aspects. Once a schedule is generated for the different courses, another problem is faced. For each course, groups have to be assigned to the different instructors in such a way that conflicts are minimized. A number of facts can generate such conflicts. For example, an instructor might be teaching different courses with overlapping schedules. In addition, each instructor has a preferable off day. Moreover, in courses with labs and tutorials, a constraint that the lab and the tutorial are to be taught by the same instructor might be needed. Therefore, the aim of this thesis is to implement a system that given inputs such as the available schedule and the instructors assignment to the different courses, would be able to generate slot assignments to the instructors taking into consideration the different constraints in such way that would minimize the conflicts. Although Constraint Programming is usually used for solving such combinatorial problems, the system can be implemented using any language.
- Portfolio management
  The financial market is changing and the need for simulation tools to select a portfolio while minimizing correlation between stocks is increasing. Portfolio selection is about finding the stock portfolio that minimizes the risk for a required return or maximizes the return for a given level of risk.
  Project Tasks:
  1. Portfolio selection Application
    
    Supervised by:
    
    Prof. Dr. Carmen Gervet
    
    the topic of this thesis is to investigate alternative optimization methods for portfolio selection. This thesis will consider well-defined and simplified forms of such models and use various constraint solving techniques to investigate the problem of portfolio selection using stocks from the Egyptian market as data and given different optimization functions. No prior knowledge of the financial market is required.
- Constraint Handling Rules
  The rule-based concurrent programming language Constraint Handling Rules (CHR) was introduced to ease the development of constraint solvers. Currently several CHR libraries exist in languages such as Prolog, Haskell and Java, worldwide more than 50 projects use CHR. CHR and dozens of its constraint solvers can be used online via the internet at WebCHR. Over time it has become apparent that CHR and its extensions are useful for implementing reasoning systems in general, including deduction and abduction, since techniques like forward and backward chaining, bottom-up and top-down evaluation, integrity constraints, tabulation can be easily implemented and combined.
  Project Tasks:
  1. Confluence of CHR programs Research
    
    Supervised by:
    
    Prof. Dr. Slim Abdennadher
    
    With CHR programs an often recurring question is whether for a certain input the calculated result is always the same, no matter in which order rules are applied. This property is called confluence and is decidable for terminating CHR programs. Within this work the student should familiarize herself with CHR and confluence. An existing automatic confluence test for CHR programs should be extended to support invariant-based confluence testing. Required knowledge: knowledge about Prolog. Knowledge about CHR is advantageous.
  2. A Java Visualization Tool for Transformed CHR Programs Research
    
    Supervised by:
    
    Prof. Dr. Slim Abdennadher
    
    Eng. Nada Sharaf
    
    The CHR research community often prototypes possible extensions of CHR by means of source-to-source transformations, i.e. new features of the language are realized by providing implementations based on the core CHR language. A new source-to-source transformation tool for CHR was already implemented. The tool aimed at adding visualization features for CHR programs. Java visualization tools were implemented for: • tracing the execution of CHR programs. • visualizing CHR constraints as geometric objects. • visualizing the execution of different algorithms such as: tree-searching algorithms, sorting algorithms and Boolean-algebra algorithms. The aim of the thesis is to implement new Java visualization tools for different visualization purposes such as visualizing Graph algorithms for example.

Developing Apps for Intel Ultrabook and Win8 Platforms

In the past, every PC had a keyboard and a mouse. User was able to sit and focus 100% on the screen. Interfaces were complex, like MS Office. Nowadays, users are always moving : they need touch screen or movement sensors to replace the keyboard. They do a lot of things concurrently, forcing interfaces to be a lot simpler and intuitive. That’s why Intel invented the Ultrabook and Ultrabook-convertibles, and Microsoft invented Win8 and metro.

Related Bachelor Projects:
- Innovative SW Usage Model
  As a developer, you’ll need to change the way you think about user interfaces and software usage models. You’ll need to optimize the power consumption of your software, about using multiple cores with OpenCL. Check the websites for more information : http://software.intel.com/en-us/ultrabook/ http://msdn.microsoft.com/en-us/windows/apps/br229516
  Project Tasks:
  1. Innovative SW Usage Model Application
    
    Supervised by:
    
    Prof. Dr. Ayman Elnaggar
    
    The ideal software will : 1. Work fine on a ultrabook with keyboard and mouse, then will have a simpler mode of operation when the ultrabook is converted to tablet mode, adapted to multitouch 2. will interact well with the power saving scheme of Win8 3. if the software is executing intensive computation (like image processing), it will use OpenCL hybrid computing
    Pre-requisite courses:
    
    CSEN702: Microprocessors

Electrocardiogram (ECG) Processing and Analysis

An electrocardiogram (ECG) is an electrical signal representing the cardiac activity of the heart. This signal is generally recorded by means a certain number of electtrodes, properly pasted on the body. ECG is most widely used signal in clinical practice as it conveys information regarding the electrical function of the heart, by altering the shape of its constituent waves. Different operations can be performed on long records of ECGs in order to filter them out, compress records to save storage space or to send over communication channels, extract pathological information, or to model the signals for further processing. MIT-BIH ECG database will be used in the projects below.

Related Bachelor Projects:
- ECG Modeling
  The shape of the ECG beat can dynamically change and is highly correlated with type of pathology. Modeling the ECG is of great importance, since the mathematical model can be used in different ways. For instance, the model can be employed for removing noise from the ECG beat in case of low signal to noise ratio.
  Project Tasks:
  1. Statistical and Emerging ECG Modeling Research
    
    Supervised by:
    
    Prof. Dr. Hazem Abbas
    
    The project deals with using different modeling algorithms that are either statistical or based on emerging methods such as neural and fuzzy methods. It also aims at comparing their ability of these methods to represent different arraythmia which can be implemented in automated ICU diagnosis.
    Pre-requisite courses:
    
    CSEN301: Data Structures and Algorithms
    
    COMM602: Digital Signal Processing
- ECG Redundancy Reduction
  ECG data compression is gaining importance in biomedical engineering due to the potential applications in telemedicine. The Holter monitoring system, for example, consists of obtaining cardiac signals over 24 h, using from 5 to 7 electrodes. Consequently, the storage of such signals is an important subject on the implementation of ECG acquisition systems. Also the transmission of the obtained signals is becoming more crucial with the growing advance of the telemedicine.
  Project Tasks:
  1. ECG Compression Research
    
    Supervised by:
    
    Prof. Dr. Hazem Abbas
    
    The task explores different ECG compression techniques that are based on computational intelligence paradigms such as neural nets and genetic algorithms and advise on particular techniques for different types of arrythmia.
    Pre-requisite courses:
    
    CSEN301: Data Structures and Algorithms
    
    COMM602: Digital Signal Processing
- ECG Arrythmia Classification
  ECG Arrhythmia is diagnosed through an ECG procedure. Because arrhythmia represents abrupt and abnormal ECG beats, physicians diagnose arrhythmia based on longterm ECG data using an ECG recording system like the Holter recorder.
  Project Tasks:
  1. ECG Classification and Analysis Research
    
    Supervised by:
    
    Prof. Dr. Hazem Abbas
    
    The importance of an automatic arrhythmia classification algorithm is being increasingly recognized. The task focuses on applying arrhythmia classification algorithms using pattern recognition and machine learning algorithms to provide fast and accurate performance.
    Pre-requisite courses:
    
    CSEN301: Data Structures and Algorithms
    
    COMM602: Digital Signal Processing

Game with a Purpose and Crowd Sourcing

Game with a Purpose (GWAP) make use of the contributions submitted by players of different Web gaming platforms in order to collect data. The games are designed in such a way so that while users are playing, needed data is collected. Crowdsourcing is an emerging approach that involves humans in solving problems which, so far, have no pure algorithmic solutions. Both fields are increasingly gaining importance. They aim at making use of human intelligence.

Related Bachelor Projects:
- Platforms for GWAP and Crowdsourcing
  In the research fields of GWAP and Crowdsourcing, several issues are investigated. Such issues are primarily related to motivating humans to contribute to some common tasks. Incentives are investigated from a theoretical as well as from a practical view point. With GWAP and Crowdsourcing platforms, users may participate in solving problems without necessarily being fully aware of the purpose they are contributing to.
  Project Tasks:
  1. Egyptian Monument Locator Research
    
    Supervised by:
    
    Prof. Dr. Slim Abdennadher
    
    Given an Egyptian monument/landmark, we need to find its geolocation (latitude/longitude) on the map. You can assume that you are given the accurate name of the object. A suggested approach is to check APIs such as those of Google Maps, and Google Earth. Lest the system was unable to locate the object (because it is too small or too specific to show on the maps), it should ask the user for a hint of its location, the hint can be as general as the city’s name (e.g. Giza) or as specific as a museum’s name (e.g. Egyptian Museum). The system should accordingly crawl other websites (e.g. Wikipedia) in order to determine a more specific location of that object, and reflect it on the map. Finally, if the application was totally unable to locate the item, it should ask the user to point to its location on the map, and save this as a possible answer for future queries. Required skills: • Web development skills • Both front-end and back-end development skills • Previous experience with APIs is a big plus
  2. Egyptian Dialect to Standard Arabic Research
    
    Supervised by:
    
    Prof. Dr. Slim Abdennadher
    
    Depending on the region in the Arab world, Arabs speak different dialects of Arabic. The purpose of this thesis is to collect a large database of words in non-standard dialects, that correspond to standard Arabic words. In the end, the collected data should be exposed in a queryable way. The data can be collected in many ways, the student can undertake either of the two suggestions listed below, or suggest a different innovative approach (after approving it from the supervisor). Approach 1: The suggested approach is to introduce a Game With A Purpose (GWAP), that asks players to provide words/sentences in their dialects that correspond to their equivalents in standard Arabic. Several players would be grouped in the same game, and get higher scores the more similar the inputs they provide are. Approach 2: The second approach A crowd-sourced website where users would provide corresponding words/sentences in their dialects that correspond to their equivalents in standard Arabic. There should be an innovative incentive that would motivate users to do that, relying solely on the users’ altruism is discouraged. Required skills: • Strong web development skills • Strong Knowledge of front-end development technologies: HTML5, CSS3, Javascript, JQuery and JQuery UI • Previous experience with building web games is a big plus
  3. Overview of Image Recognition Tools for Egyptian Landmarks Application
    
    Supervised by:
    
    Prof. Dr. Slim Abdennadher
    
    Mohamed Khamis
    
    There are many services that provide image recognition services, perhaps the most famous of them are Google Goggles and Google Images. The purpose of this thesis is to investigate and find a service suitable for recognizing landmarks. The idea is that tourists would be able to provide the pictures they took during their sight-seeings, and the service would understand which Egyptian landmark the tourist took a picture of. The thesis work will mainly involve examining different APIs and analyzing their performance to identify the most suitable approach. The approach could be as simple as using an existing API, or combining the results from different ones to achieve better results. By the end of the thesis, a tool should be implemented (a website for example) to demonstrate using the approach that proved most suitable for the task. The tool should ask tourists to provide pictures of the landmarks, and reply back with some information about the landmark, as well as similar images that were taken by other tourists. P.S. This thesis does not involve image processing; it aims to rather make use of the existing technologies in this field. Required skills: • Dealt with APIs and web services before • Web/mobile development skills
  4. ARTigo 2.0 - Identifying Objects Research
    
    Supervised by:
    
    Prof. Dr. Slim Abdennadher
    
    ARTigo is a Game With A Purpose (GWAP) to collect tags for paintings. Players earn higher scores if their inputs match with inputs of other players. The game was successful for collecting tags for many paintings. A problem is that images are given tags that correspond to objects within the image even if they were of low significance. For example, a painting showing people having dinner may contain an “Apple”. Accordingly, the tag “Apple” is acceptable for that painting, although it is not the main theme that the painting revolves about. The purpose of this project is to further identify objects in the paintings by tagging the pixels and not just the image. A suggested approach is to use the results collected from the initial version of ARTigo. Two players would be selected at random, one of them assigned to be a describer and another to be a guesser. The describer is given a painting and one of the previously collected tags, and would try to choose some pixels to show to the guesser. If the guesser, who is only able to see the pixels the describer selected, was able to identify the tag, we can deduce that the uncovered pixels sufficiently describe the tag, and thus we can label these pixels with the tag. For example, a painting showing a plate of fruits can have a tag “Apple”. The describer, given the tag “Apple” and the painting, should choose the pixels from the painting that he thinks best describe the “Apple”, they would both earn points if the guesser guessed the word “Apple”. The results from the game can be used to enhance search engines. The improved search engine could respond to a query “Apple” by showing the painting that contains the object, and pointing at it. Additionally, search engines could deduce the significance of the object within the image by measuring the ratio between the pixels consumed by the object to the total number of pixels in the image.
  5. AreCAPTCHA - Arabic reCAPTCHA Research
    
    Supervised by:
    
    Prof. Dr. Slim Abdennadher
    
    Mohamed Khamis
    
    Millions of internet-based services are available online nowadays, and thousands are introduced every day. Most, if not all, of these services rely heavily on web-forms, where users are expected to input text that is later on processed by servers. Nevertheless, it is easy for many internet users to build programs that can automatically fill these forms, and submit them abusively many times. We refer to these programs as “bots”. The reason behind this vulnerability is that bots can easily understand the expected input in the different fields. A counter measure, called CAPTCHA, was proposed by von Ahn [1]. CAPTCHA is a program that can automatically generate and grade tests that are easily solvable by humans, but not by computer programs. This is done by asking the user to identify some distorted characters, that are too difficult for Optical Character Recognition (OCR) programs to read. CAPTCHAs proved successful, and are widely used in most, if not all, online web services. A second version of CAPTCHA is called reCAPTCHA. ReCAPTCHA is the first application of Human Computation, a field that aims to utilize the human brain power that is used over the internet, by channeling tasks that can easily be solved by humans. The newer version shows two words instead of random characters, one of the words is known by the server, and this is the one used to validate that the entity filing the form is a human, while the other word is unknown by the server. The other word is actually scanned from a book that was never digitized, because advanced OCR programs were not able to read them. Eventually, if a large number of users, say 100 users, agree that a certain scanned word is the word “Hello”, then it is stored as a possible digitization of this word. By doing this, humans over the internet contribute in digitizing books while filling the online web forms. ReCAPTCHA has been widely successful. Currently almost 100 million reCAPTCHAs are filled every day. The technology was able to digitize 130 years of the New York Times magazine in 12 months. Von Ahn founded reCAPTCHA as a startup, and it was later acquired by Google, who uses it now for one of its core products: Google Books. We propose an equivalent system for digitizing Arabic text. Unlike hard-copy books, digital books do not lose quality over time. Egypt has lost many important books in recent events, digitizing these books using the reCAPTCHA concept will consume less resources than those needed to do the task by paid workers. If a worker is paid a fraction of one EGP to digitize a page, digitizing ten 500 pages books will cost a lot. On the other hand, there are many Arabic websites that use English reCAPTCHAs, although the majority of their users are Arabic speakers. It could be negotiated with these website owners to replace their English reCAPTCHAs with AreCAPTCHA, while providing a safe fall-back option, to assure the website administrators that their safety measure will never fail. The proposed project would have the following objectives: • Develop a methodology to extract the words that are unreadable by OCR programs • Develop the AreCAPTCHA web service that gives the user two words: the word to be digitized, and another to serve as a human test. The service should be easily integrated into websites to compete with the existing reCAPTCHA by von Ahn. • Maintain an easy way to switch to the regular English reCAPTCHA should be available • Determine a threshold for the number of times a word needs to be digitized. • Perform a final experiment to test AreCAPTCHA.

GUC Software Foundation

Projects involved in the "GUC Software Foundation" are mainly applications being developed to help either the GUC develop its software infrastructure or to help students to understand the concept taught in different courses.

Related Bachelor Projects:
- Teaching tool support
  Teaching and student interaction depends on a large number of menial tasks, such as material preparation, setting up of assignments, quizzes, exercises, and exams, tractability of example/exercise problems, correct weighting of problems towards the course objectives and difficulty level, etc. Further, electronically available examples, exercise and experimentation opportunities, and tutorials promise to be of tremendous help specifically within the mature fields of basic teaching. In this project, various tools for in classroom and off classroom teaching and administrative work will be created. Their creation will require creativity, good overview to understand complex, but repetitive tasks, and a good sense for useful and interesting experimentation.
  Project Tasks:
  1. Pseudo-code experimental interpreter Application
    
    Supervised by:
    
    Assoc. Prof. Georg Jung
    
    One of the biggest hurdles for GUC engineering students in their first semester is to understand the mechanics of simple algorithms. The learning curve of an actual programming language such as Java is steep, so it is not suitable for starting to teach the concepts of algorithms. Pseudo code on the other hand is abstract, no execution environment is available, and the only feedback the student receives is the marks on his or her assignments. To really understand the ideas of algorithms, experimentation with immediate feedback is extremely helpful. In this task, an interpreter for a modified pseudo-code will be built, that allows for entering, editing, and step-by-step executing of mini-code while observing the effects of statements and control-flow as they happen. Understanding the minimum basics of compiler construction will be very helpful. Not a prerequisite but also helpful would be some knowledge from CSEN403 - Concepts of programming languages.
    Pre-requisite courses:
    
    CSEN1003: Compiler
  2. Courses manager Application
    
    Supervised by:
    
    Assoc. Prof. Georg Jung
    
    When planning a lecture, the lecturer has to consolidate multiple parallel aims. First, the learning outcomes and course objectives have to be categorized and sorted according to a learning taxonomy. The material has to be structured and weighted and distributed over a given semester schedule. According to the weighting of each topic, practice assignments, assignments, quizzes and exams have to be generated from a well managed body of general as well as topic-specific problems. To comply with regulations, the ratio of exercises covering specific topics in comparison to other topics within exams has to correspond to the importance of the topic within the course objectives as reflected by the fraction of time that was spent on the topic during the semester. The task is to develop an integreated special purpose development environment that supports planning and managing of courses, including and integrated exercise database and collection of statistics and feedback.
  3. Major selection guide for engineering students Application
    
    Supervised by:
    
    Prof. Dr. Carmen Gervet
    
    The student will derive a prototype tool that will guide engineering students to choose their Engineering major by defining a set of criteria to assign preferences to (variable), such that the tool should derive/infer a preferred major, less preferred and definitely not desired major for instance. The tool should be integrated on the MET website.
  4. Student course evaluation Application
    
    Supervised by:
    
    Prof. Dr. Carmen Gervet
    
    The student will build a prototype tool that will query the user on a set of criteria, take into account student attendance, etc.. and seek appropriate measures of evaluation of a given course. The idea is to simulate with different measures and seek "reliability".
  5. Student semester load simulator Application
    
    Supervised by:
    
    Prof. Dr. Carmen Gervet
    
    The student will derive a prototype tool, that will take the schedule per semester class for MET students and compute the estimated work load overall. He/she will also derive (in case of unbalanced work loads) potential solutions to redistribute the load and courses taking into account prerequisites.
  6. Matching GUC majors with Alumni jobs Application
    
    Supervised by:
    
    Prof. Dr. Carmen Gervet
    
    In this thesis the student will define a graph of dependencies between GUC majors and the different jobs/companies graduates are working in today. This includes a structure around domains, companies, then job titles. A prototype tool will be implemented and when queried one should be able to match majors with companies, titles and the duration of employment in them if possible
  7. Faculty statistical tool Application
    
    Supervised by:
    
    Prof. Dr. Carmen Gervet
    
    This project will build the architecture and prototype for a statistical tool that considers each major, and computes using different layout (tables, graphs) facilities, the statistics over the past 10 years, based on advanced criteria the student can determine. This project involves both creativity, programming, and GUI design.

Human Activity Recognition

Modern mobile devices are equipped with a number of sensors such as accelerometers, magnetic compass, and gyroscopes, which help in providing many kinds of assistance to the user. For example, these sensors can detect a fatal accident happening to a living-alone senior such as passing out after falling on a slippery floor. Such incident is manifested as a short abrupt motion followed by a relatively long period of no motion. The inertial sensors, such as accelerometers, are integrated within modern personal devices thus require less effort in terms of hardware setup and maintenance compared to vision-based activity recognition systems.

Related Bachelor Projects:
- Comparing Different Techniques for Human Activity Recognition
  The project involves comparing the performance of a number of activity recognition techniques using Matlab or Java.
  Project Tasks:
  1. Gyroscopes for Human Activity Recognition Application
    
    Supervised by:
    
    Dr. Hisham Othman
    
    Gyroscopes are inertial sensors that are used to measure the orientation of platform with respect to a given reference. Due to the recent advancements in Micro-Electro-Mechanical Systems (MEMS) manufacturing, gyroscopes are now produced in a small form and even integrated within high-end smart phones. Yet, gyroscopes are not as widely spread as their cosines accelerometers. This project aim to evaluate the value added to the human activity recognition by using gyroscopes alone or a combination of gyroscopes and accelerometers in human activity recognition applications.

Hyperspectral Images

Hyperspectral images are three-dimensional images that are composed of two spatial dimensions and one spectral dimension. They are formed by capturing the image through a relatively large number of optical filters centered at different wavelength in the visible light range, near-infra-red range and short wave infra-red range. The hyperspectral images are used applications such as remote sensing and detecting currency counterfeits.

Related Bachelor Projects:
- Denoising of Hyperspectral Images
  The purpose of this project is to reduce the sensor noise that exists as intrinsic part of the capturing process. This requires basic knowledge about color images and Matlab.
  Project Tasks:
  1. Denoising of Hyperspectral Images Application
    
    Supervised by:
    
    Dr. Hisham Othman
    
    The purpose of this project is to reduce the sensor noise that exists as intrinsic part of the capturing process. This requires basic knowledge about color images and Matlab.
- Endmember Extraction from of Hyperspectral Images
  Endmembers are a collection of a scene's constituent spectra in a hyperspectral image. Each endmember has a known spectral signature but due to the aggregation of more than one endmember in one pixel, the pixel spectral signature becomes a weighted some of a number of spectral signatures. One major task in remote sensing application is the extraction of these endmembers and providing maps of their abundances from a given hyperspectral image.
  Project Tasks:
  1. Endmember Extraction from of Hyperspectral Images Research
    
    Supervised by:
    
    Dr. Hisham Othman
    
    The purpose of this project is to investigate the limitations and the challenges of endmember extraction in low-resolution hyperspectral images. The simulations are to be run on Matlab. Basic knowledge of image processing is required.

Hypertext Framework

Hypertext is a means to structure information. There are various structure types supported by hypertext applications. The most dominant are node-link structures. The WWW is based on that paradigm. Besides that there are spatial hypertexts (i.e., associating via visual cues), taxonomic structures (i.e., classifications), argumentation supporting structures (e.g., for collaborative decision making processes), or hypertext fiction (i.e., fiction represented as hypertexts). They all come with their own structure abstractions and behavior. This creates specific requirements on a system that has the potential to support them.

Related Bachelor Projects:
- Component-based Hypertext Framework
  This project aims at developing the core of a Component-Based Open Hypermedia System (CB-OHS). This is an open infrastructure that provides access to core functionality (storage, naming, lookup services, etc.) to all of its services. Support for specific structure types (e g., node-link, spatial, taxonomic, etc.) is provided by individual structure services that can be added to or removed from the framework in runtime.
  Project Tasks:
  1. Design of the Framework and Implementing Core Services Research
    
    Supervised by:
    
    Dr. Fatma Meawad
    
    After a detailed literature survey of the state of the art, there is a discussion on design issues of the system, including finding the most appropriate system architecture (e.g., client-server or P2P) for the task at hand. The prototype should include basic services as a proof on concept.

Integrated meta-modeling based software engineering framework

Model driven development and meta-modeling is a novel approach to software engineering that focuses on radical modularization and consistent, multi-level, abstraction and concretization (i.e., development) of software artifacts through hierarchic modeling. The aims of the model driven development paradigm are (a) a better management of the enormous complexity of modern-day software, (b) better re-use of existing software in various contexts, (c) better maintainability of software systems, (d) increased reliability of software systems, and (e) a substantial decrease in the costs of software development. In this topic, CADENA, a model-based software engineering framework developed at Kansas State University will be enhanced and recreated as an Eclipse plugin. The name of the project is CADENA_e, where 'e' can stand for version 5, "enhanced", "evolution", "embedded systems", or "Egypt", whichever you like best. All projects and tasks within this topic will require you to use Java, Git, Eclipse, some teamwork, and of course LaTeX

Related Bachelor Projects:
- Structural modeling
  CADENA_e is based on a component oriented programming paradigm. Components are software artifacts that form the primary building blocks of a system (comparable to objects in object oriented programming), they are defined by their interfaces and can be easily plugged together or interchanged. Component diagrams specify how various components interact within a larger system. The structural modeling tools in CADENA_e should allow to design component diagram styles, create components, and assemble the components into scenarios.
  Project Tasks:
  1. Form-based component diagram editing feature Application
    
    Supervised by:
    
    Assoc. Prof. Georg Jung
    
    You will use the form editing features in eclipse to build a three tier component graph creation and editing feature for CADENA_e.
  2. Graph-based component diagram editing feature Application
    
    Supervised by:
    
    Assoc. Prof. Georg Jung
    
    You will use the graph editing features in eclipse to build a three tier component graph creation and editing feature for CADENA_e.
- Behavioral modeling
  The original CADENA featured only structural modeling. However, a generic, abstract, behavioral specification in form of activity graphs will mark an important step towards pure model driven development (where actual coding can be reduced to a minimum or even eliminated). I this project, graph-based and form-based editors for behavioral modeling based on lightweight process coordination will be implemented.
  Project Tasks:
  1. Graph-based process specification feature Application
    
    Supervised by:
    
    Assoc. Prof. Georg Jung
    
    You will implement a creation and editing tool for processes based on the Eclipse graph libraries and lightweight process coordination.
  2. Form-based process specification feature Application
    
    Supervised by:
    
    Assoc. Prof. Georg Jung
    
    You will use the form editing tools in Eclipse to implement an editing feature for various process algebras or similar process description formalisms.
  3. Process execution environment Application
    
    Supervised by:
    
    Assoc. Prof. Georg Jung
    
    You will animate the behavioral specifications (or even execute the various processes) in a test-run environment for validating and debugging.
- Verification/validation features
  A major advantage of model driven development is that the "code" (i.e., the models) are abstract enough to easily introduce well established formal verification procedures without much preprocessing. This project will enable model checking on the given behavioral models.
  Project Tasks:
  1. Behavioral model back-end translator Application
    
    Supervised by:
    
    Assoc. Prof. Georg Jung
    
    You will translate the given behavioral description in CADENA_e into the input syntax of a well known model-cheking tool. Infrastructure to automatically run the model checker is required, editing support or backwards translation would be a bonus.
  2. Time-edit verification formula generator Application
    
    Supervised by:
    
    Assoc. Prof. Georg Jung
    
    Some advanced input artifacts of modern lightweight formal verification tools are not easy to use for a software engineer without specific training. In this task, you will break down the generation of LTL formulas to simple time-line manipulations that are intuitive to use. Knowledge from CSEN1017 is helpful, but not required.
- Dimensional typing
  Sensors measure physical sizes or quantities. These quantities are generally compromised of a scalar value and a unit. The units, in turn, are related to each other by their physical dimension. For example, length is measured in meters, so when measuring a distance the result would be a scalar X and the unit Meter (Xm). Speed for example is the distance per time unit (the first derivation of length over time). Therefore, the dimension of speed is length divided by time. Likewise, the dimension of acceleration is speed divided by time, or length by time square. The source of some of the most frequently occurring errors in the usage of sensors and unit-carrying values are conversion errors that could be avoided by the proven technique of typing. This project aims to create a dimensional typing system for a plug and play sensor service and component system.
  Project Tasks:
  1. Dimensional type system Application
    
    Supervised by:
    
    Assoc. Prof. Georg Jung
    
    You will develop and implement a dimensional type system for physical quantities. It should allow assignments such as "speed = (1km + 25m) / 20s".
- Tool support
  Some features projected for CADENA_e are independent from the conceptual part and mainly serve editing and visualization convenience. In this project, such features will be implemented.
  Project Tasks:
  1. Multi-feature graph layouter Application
    
    Supervised by:
    
    Assoc. Prof. Georg Jung
    
    Graph-based editing of complex systems often suffers from the size of the resulting graphs and the lack of structure and overview. In this task, you will implement a specialized graph layouter based on various standard graph layouting algorithms with added features such as subgraph-folding, fixing of specific nodes, sorting of nodes according to specific properties, etc.
- Case study
  We will build an example and experimentation system that will gradually include the ideas of model driven development for testing the concepts of CADENA_e in practice.
  Project Tasks:
  1. Real-time fitness monitor Application
    
    Supervised by:
    
    Assoc. Prof. Georg Jung
    
    You will implement a multi-sensor tracking and analysis tool for a specific physical workout routine to track performance and fitness level of a practitioner over a single and over multiple sessions.
  2. Two student-proposed topics Application
    
    Supervised by:
    
    Assoc. Prof. Georg Jung
    
    [only one topic free, availability of Microsoft Kinect system prerequisite]

Intelligent Web Applications

Modern web applications are no longer just a portal to surf data. This is present in our daily use of the internet. Whether that is through searching via Google, using suggestions from Facebook or Twitter, or even considering suggested items on Amazon, web applications are no longer just a place to store and view data. The involvement of Machine Learning, Data Mining, and Natural Language Processing techniques helps make use of data to obtain interesting observations and relations to offer to the user.

Related Bachelor Projects:
- Educational Tools
  In an educational institute, students are faced with frequent tasks and assignments that they need to submit to their instructor for assessment. On the other hand, instructors need to provide content to the students, along with managing the material submitted by students.
  Project Tasks:
  1. Question Bank Application
    
    Supervised by:
    
    Prof. Dr. Slim Abdennadher
    
    Course instructors are often faced with the challenge of writing questions for assignments, quizzes, and exams. Therefore, causes a hassle to make sure that questions are not repeated. The aim of this project is to create a web application that would help instructors maintain the questions for any course. Instructors should be further able to generate different evaluation methods (quizzes, assignments, ...) by choosing the need question from the web application directly.
  2. Lecture Feedback – backend Research
    
    Supervised by:
    
    Dr. Wael Abouelsaadat
    
    During a lecture, an instructor often glances at the students' faces to guess if they are lost or are following the material being presented! This project involves constructing a more straight forward way to communicate the students' reaction to the material being presented. The goal is to build a website to enable students to provide instantaneous feedback during a lecture. The instructor uploads the slides to the website before the lecture. Students logon during the lecture and using a friendly interface communicate their general understanding of each slide using a simple scale; for example; following, lost, not sure I understand, day dreaming, etc... The instructor can also logon to the website during the lecture to see the overall impression of the class to each slide. The website will also enable the instructor to conduct anonymous quizzes to check which parts are unclear. This part of the project involves building the backend database and logic layer supporting the website. During this project, you will have a chance to train in large scale website development, and various backend technologies.
  3. Lecture Feedback – front end Research
    
    Supervised by:
    
    Dr. Wael Abouelsaadat
    
    During a lecture, an instructor often glances at the students' faces to guess if they are lost or are following the material being presented! This project involves constructing a more straight forward way to communicate the students' reaction to the material being presented. The goal is to build a website to enable students to provide instantaneous feedback during a lecture. The instructor uploads the slides to the website before the lecture. Students logon during the lecture and communicate their general understanding of each slide using a simple scale; for example; following, lost, not sure I understand, day dreaming, etc... The instructor can also logon to the website during the lecture to see the overall impression of the class to each slide. The website will also enable the instructor to conduct anonymous quizzes to check which parts are unclear. This part of the project involves building the web front end. You will have a chance to train in various advanced UI dev techniques including RIA, Ajax, and using third party libraries.

Java SNePS

SNePS is one of the oldest, yet developing, systems for knoweldge representation and reasoning in AI. Akin to a programming language, SNePS is a flexible system for building and manipulating propositional semantic networks. A propositional semantic network is a labelled directed graph were nodes represent entities (possibly propositions) and arcs represent structural relations among them. The basic SNePS system is a network managing system concerned with building and finding nodes. On top of that, there are several sub-systems for reasoning, acting, belief-revision, and natural language understanding and generation. The current version of SNePS is implemented in Common Lisp. In a series of projects, we shall attempt to implement a Java version of the complete SNePS system.

Related Bachelor Projects:
- SNeRE
  SNeRE, the SNePS Rational Engine, is the component of SNePS responsible for inference and acting. Both inference and acting are interwined, following the slogan "Acting in the service of inference, and inference in the service of acting". The design of SNeRE is based on a network of processes each associated with a node. Processes run concurrently and communicate and synchronize through message passing under the control of the SNeRE executive.
  Project Tasks:
  1. Cleaning Up Java SNIP Application
    
    Supervised by:
    
    Dr. Haythem Ismail
    
    One student will finalize the implementation of Java SNIP---the Java SNePS inference package. For example, the full integration of SNIP with SNeBR (the SNePS belief revision system) is needed and the inference cycle needs to be completed.
  2. Default Reasoning in Java SNePS Application
    
    Supervised by:
    
    Dr. Haythem Ismail
    
    Default reasoning is a form of nonmonotonic reasoning that allows humans to jump to reasonable conclusions. The assumptions underlying these "jumps" are called "default rules". For example, on learning that someone is a high-school student, one immediately places the person within a certain age group. Such a move is only justified by a default rule, not by any kind of logical necessity. One student will implement default reasnoing in Java SNePS. The implementation will be guided by the work done by Bhushan (2003), where, by default, all SNePS rules are considered default rules. This admits conflicting rules, but a global ordering of rules resolves conflicts in favor of the more preferred defaults.
  3. Rule Introduction in Java SNIP Application
    
    Supervised by:
    
    Dr. Haythem Ismail
    
    The Java SNePS inference package (SNIP) is a process model impementation of a natural deduction system. In the current implementation, only elimination natural deduction rules exist; introduction rules have not yet been implemented. An introduction rule is a rule that allows the system to infer a new compound sentence. For example, to infer the rule P => Q, the system may assume P and try to infer Q. One student will explore rule introduction in Lisp SNePS and start its design and implementation in Java SNePS.
- SNePS Language Interfaces
  Interacting with SNePS is facilitated by three interface languages: (i) SNePSUL, a Lisp-like user language, (ii) SNePSLOG, a first-order-like logical language, and (iii) a fragment of English defined by a user-defined grammar and lexicon.
  Project Tasks:
  1. SNePSLOG Application
    
    Supervised by:
    
    Dr. Haythem Ismail
    
    One student will implement the SNePSLOG interface to the current Java SNePS system. SNePSLOG is a classical-logic-like language allowing users to interact with the SNePS system by entering SNePSLOG commands. The interface should be user frindly and should correctly impelment all SNePSLOG commands.
- Matching
  Unification is the heart of formal first-order reasoning. Two first-order expressions unify if there is a variable substitution which, when applied on them, yields identical expressions. Unification in SNePS is complicated by path-based pattern matching. (This is based on user-defined equivalence relations between paths and single relations.)
  Project Tasks:
  1. Matching in Java SNePS Application
    
    Supervised by:
    
    Dr. Haythem Ismail
    
    One student will explore matching in Lisp SNePS and amend (and possibly redesign) the current implementation of matching in Java SNePS.

Logic of Knowledge

Propositional knowledge is the type of knowledge expressed by sentences of the form "A knows that p", where A is an agent and p a proposition (statement). A classical analysis holds that propositional knowledge is justified true belief. Such an analysis has come under attack and has, thenceforth, been refined. Unfortunately, formal logics of knowledge have not yet matured enough to mirror the advances in philosophical analysis (with the possible exception of the so-called justification logic). This is unfortunate since formal logics of knowledge have had diverse applications in multi-agent systems and reasoning about network security, to mention only two application areas. This gap needs to be filled.

Related Bachelor Projects:
- The Non-Monotonic Logic of Justified True Belief
  The objective of this project is to develop a logic of knowledge that captures recent amendments to the thesis of knowledge as justified true belief. It is believed that for such a notion of knowledge to be put into application, a non-monotonic logic is called for.
  Project Tasks:
  1. An Investigation into the Logic of Knowledge Research
    
    Supervised by:
    
    Dr. Haythem Ismail
    
    One student will review the literature on the analysis of knowledge and the literature on the logic of knowledge. The outcome should be an evaluation of the various logics of knowledge proposed with respect to how much they capture current analyses of knowledge.

Orientation Sensors

The automatic control in some systems requires knowledge of the orientation of the system in the three-dimensional space. This knowledge can be acquired by a set of sensors including accelerometers, gyroscopes and magnetic sensors. The latter provides the orientation with respect to the direction of the earth magnetic pole depending on the magnetic field of the earth. Unfortunately, there exists in many cases an interference from stray magnetic fields that may result from nearby electrical motors, electricity power transmission lines, or electromagnetic waves transmitters.

Related Bachelor Projects:
- The Effect of Stray Magnetic Fields on Orientation Sensors
  The purpose of this project is to investigating the effect of stray magnetic fields on the performance of the orientation sensors.
  Project Tasks:
  1. The Effect of Stray Magnetic Fields on Orientation Sensors Research
    
    Supervised by:
    
    Dr. Hisham Othman
    
    The student is expected to carry out simulations using Matlab or Java and analyze the resulting outcomes.

Programming Language Localization

Language is currently a barrier for teaching programming to children of non-English speaking countries. Hence, a new research direction aims to provide a localized version of programming languages. For example, Yijun Yu & Mylopoulos proposed using a localized version of XML on the web.

Related Bachelor Projects:
- IDE-Based Localization
  One approach to localizing a programming language is to add the localization to the IDE itself. This approach has the advantage that if the IDE supports more than one language, then the same localization module could be reused.
  Project Tasks:
  1. Eclipse PL Localization Research
    
    Supervised by:
    
    Dr. Wael Abouelsaadat
    
    In this project, you are going to modify Eclipse source code to add a translation layer. Using the translation layer, a user can define corresponding keywords and reserved words of a specific programming language (PL) in his own language. Thus an Egyptian kid can write a program in Arabic and a German kid can write it in German. The translation layer will translate the keywords, reserved words and user identifiers back to English. When a user attempts to compile his/her program, the translation layer translates the compilation massages back to the programmer's own native language. During this project, you will be exposed to Eclipse source code, a large app with sophisticated architecture and extensive use of design patterns. You will also have a chance to learn about scanners and parsers.

RDBMS

Databases are complex programs, with competing memory and memory-footprint requirements. Along with operating systems, they are considered the category of apps to learn about writing sophisticated and large applications. However, most open source RDBMS are too complex to be studied at the undergraduate level (e.g. MySQL is 1.5Mn LOC).

Related Bachelor Projects:
- Educational RDBMS
  This project involves building a mini RDMBS that could be used by other students around the world to learn about RDBMS internals. The engine will be designed following exo-kernel principles, to allow for other developers to replace any module of the RDBMS with one of their choice.
  Project Tasks:
  1. Educational RDBMS Engine Research
    
    Supervised by:
    
    Dr. Wael Abouelsaadat
    
    This will be an opportunity for you to gain a hands-on experience in data structures, algorithms, and design patterns that are often used in building a large app. You will be responsible for designing the engine and implementing at least one module for each subsystem to have a running version by end of term. Using third party libraries is allowed and encouraged to expedite the development process.

Smart Building and simulation

This topic aims at studying the migration towards smart environment by developing simulation tools that study the minimization of operational costs meaning migrating towards automated tools that minimize people's interaction with the building facility.

Related Bachelor Projects:
- Decision support tools
  This project studies various decision problems for which graph and search algorithms are needed to improve the usage of energies in buildings.
  Project Tasks:
  1. Simulation tool for 3D Sensor placement in GUC C6-C7 corridors Application
    
    Supervised by:
    
    Prof. Dr. Carmen Gervet
    
    Consider the geometrical layout of the GUC C6-C7 corridors and rooms. In this bachelor, the student will determine the layout of light and heat sensors (virtual ones) such that the whole space can be covered to carry out measurements in a real-life setting. This task comprises two main elements: a graphic simulation of the C6-C7 corridors and an optimized placement of two types of sensors that guarantee that the whole physical 3D space is covered.

Smart Mobile Application

With the growing capabilities of mobile devices and their wide spread in the society, the potential of employing such devices towards social development opportunities is highly recognized. Adding to this the availability of data through social networks and other data APIs, promises the possibility of adding intelligence to mobile applications. The tasks added to this topic will be strongly related to this generally acknowledged opportunities. The announced ideas under this topic will be offered to applied arts students to contribute in the design part of the projects through their own bachelor projects. Additionally, the proposed ideas will aim to apply for the Etisalat Mobile competition.

Related Bachelor Projects:
- An Augmented Reality Translator - Towards a Universal Language
  The aim behind this project is to create an augmented reality application for smartphones that translates natural languages using Latin typefaces (English as a start) to pictographic symbols. The project is under a larger on-going project in Plymouth University that investigates the creation of a universal pictographic language. The project aims to find out if the usage of pictographs in a world wide language would bridge the gap between cultures resulting in a multicultural global community. Furthermore, research is running under the field of Emotional Design to study the effect of having a visual language on the psychology of the human being and whether or not such a pictographic language in some contexts would rather enhance the emotional experience of the persons engaged in the communication, and this would thus enhance the quality of communication process as a whole. Therefore, as a step in the research, this project’s goal is to develop an application that would serve as the person’s visual translation tool. Using a database of designed pictographs -each conveying a certain meaning- the smart phone application should recognize the incoming English text –in words or sentences- and translate it into the corresponding stored pictographic symbol(s). An even further investigation in the research is the aspect of “time” within a pictographic symbol. Motion in general as opposed to still life has the ability to reach the viewer on another plane of reality. And thus, has a stronger impact on the human side. Therefore, it is an interesting line of thoughThe aim behind this project is to create an augmented reality application for smartphones that translates natural languages using Latin typefaces (English as a start) to pictographic symbols. The project is under a larger on-going project that investigates the creation of a universal pictographic language. The project aims to find out if the usage of pictographs in a world wide language would bridge the gap between cultures resulting in a multicultural global community. Furthermore, research is running under the field of Emotional Design to study the effect of having a visual language on the psychology of the human being and whether or not such a pictographic language in some contexts would rather enhance the emotional experience of the persons engaged in the communication, and this would thus enhance the quality of communication process as a whole.
  Project Tasks:
  1. Augmented Reality Translator Application
    
    As a step in the research, this project’s goal is to develop an application that would serve as the person’s visual translation tool. Using a database of designed pictographs -each conveying a certain meaning- the smart phone application should recognize the incoming English text –in words or sentences- and translate it into the corresponding stored pictographic symbol(s). An even further investigation in the research is the aspect of “time” within a pictographic symbol. Motion in general as opposed to still life has the ability to reach the viewer on another plane of reality. And thus, has a stronger impact on the human side. Therefore, it is an interesting line of thought for possible integration of motion within the pictographs of our communication system. The student will work with an augmented reality SDK called vuforia and will collaborate with a designer and the PhD student working on this project at Plymouth University.
  2. Always-on Video Recording Research
    
    Supervised by:
    
    Dr. Wael Abouelsaadat
    
    This project involves building a cell phone application to continuously record a video file of a fixed length/time along with GPS location. At the end of selected time duration, recording start over from beginning of file. When something of interest happens, the user will ask the app to save the video that was recorded in a separate file. The goal is to avoid missing any interesting scene or event. For example, the cell phone could be mounted on the car's wind-chill. In case of a hit-and-run or bad driving, all the user does is to choose to save the file. The app will also extract faces and car plates from the recorded video. This project will train you in cell phone app development. In addition, you will be learning and experimenting with computer vision algorithms to choose the best set of algorithms for the task.
- Smart Taskinator
  Any large community with a complex structure suffers in the process of scheduling and assigning tasks for their members. Additionally, the need to track members’ performance and general achievements evolve; yet, it is never well planned for. The aim of this application is to build a smart application that will allow any large community to arrange themselves in a flexible hierarchy of groups with the possibility of scheduling and visualising various types of tasks of the community members and track members’ performance. The student will use machine learning to create an intelligent backend that understands the users and recommend improvements to their schedule. The requirements of this project will be provided by existing NGOs to support their current activities.
  Project Tasks:
  1. Taskinator Application
    
    This project will have a mobile side and a server side. The mobile side will be built using phonegap and mobile web technologies to cover many mobile platforms. The server side will employ several techniques for learning and tracking user activities in order to recommend smarter schedules and plans for activities. There will be extensive use and creation of webservices and online APIs.
- Mobile Payment
  Mobile payment applications are envisaged to replace money in the future. Such thought of mobile money or virtual money transfer through mobile devices is largely growing around the world especially in places where banks are scarce and where people don't normally have a bank account. The company Shoghlanah will provide the requirements of this specific project. Thus the student will be working with business analysts and consultants from a real company environment.
  Project Tasks:
  1. Mobile payment application Application
    
    The student will work with either native application or mobile web application that will implement full payment management using QR codes, currency management, card management and budget management modules. The proposed application is very rich with features and has a great future in the near future.
- Friendly indoor markers
  As part of a large Fund of a mobile augmented reality, there is an ongoing research work to support friendly markers indoors. Friendly markers can be hand drawn by users and instantly used by any mobile user. They are similar in simplicity to post-its that we use to leave a note somwhere. The aim is to create a useful office assistant using friendly markers. Imagine a Dr wanting to leave their office for a coffee and wants to leave a note on their door. Imagine that instead of using a post it, they can just go online and leave a note that can be seen by any one passing by their door, using a mobile. In this case, the marker can be the Dr.'s name or card that is already on the door.
  Project Tasks:
  1. Mobile Marker Application
    
    The student will work with android development and use the existing work of a master researcher who already developed friendly markers. the student will develop an online server side application that communicates with the mobile application and simulates the office assistant application. users can create markers and associate data to it.
- Media Applications
  In such projects, the aim is develop applications to deal with different media content such as audio, video and pictures.
  Project Tasks:
  1. Always-on Video Recording Research
    
    Supervised by:
    
    Dr. Wael Abouelsaadat
    
    This project involves building a cell phone application to continuously record a video file of a fixed length/time along with GPS location. At the end of selected time duration, recording start over from beginning of file. When something of interest happens, the user will ask the app to save the video that was recorded in a separate file. The goal is to avoid missing any interesting scene or event. For example, the cell phone could be mounted on the car's wind-chill. In case of a hit-and-run or bad driving, all the user does is to choose to save the file. The app will also extract faces and car plates from the recorded video. This project will train you in cell phone app development. In addition, you will be learning and experimenting with computer vision algorithms to choose the best set of algorithms for the task.
  2. Flower Identification Research
    
    Supervised by:
    
    Dr. Wael Abouelsaadat
    
    This project involves building a cell phone application to take a picture of a flower or a rose and identify its name and type. The application will maintain an embedded database of images to compare against. This project will train you in a cell phone app development. In addition, you will be learning and experimenting with computer vision algorithms to choose the best set of algorithms for the task.

Video Over Wireless

Develop Several Applications for Video Streaming in the very challenging limited BW of the wireless domain.

Related Bachelor Projects:
- Video Streaming Analyzer
  This project involves the development of a video streaming server and video playback client. The streaming server should support HTTP streaming format and the video streaming client should also support the same format but also it should support some of the following adaptive streaming formats 1. Microsoft smooth streaming 2. Apple Live streaming 3. Adobe Dynamic Streaming 4. DASH streaming protocol
  Project Tasks:
  1. Develop Video Streaming Analyzer Application
    
    Supervised by:
    
    Prof. Dr. Ayman Elnaggar
    
    Develop and application that parses transmitted video stream in a typical network and analyzes important statistics as shown in Figure. This application should intercept transmitted video packets, analyze them, and report statistical information based on the analysis. Some existing product can be found in http://www.telchemy.com/vqmon_streaming.php
    Pre-requisite courses:
    
    DMET703: Video and Audio Technology
- Video Content Authoring Tools
  The purpose of this tool is to open an encoded video file (.avi, or mp4,..etc.), or a separate audio/video files streaming and process them together (combining or integrating them). Students are allowed to use public media codec libraries for encoding/decoding/multiplexing for streams. However, student will also need to add more features such as GUI, multiple encoding. This project should develop an application similar to Microsoft Expression Encoder tool.
  Project Tasks:
  1. The Development of Video Content Authoring Tools Application
    
    Supervised by:
    
    Prof. Dr. Ayman Elnaggar
    
    The purpose of this tool is to open an encoded video file (.avi, or mp4,..etc.), or a separate audio/video files streaming and process them together (combining or integrating them). Students are allowed to use public media codec libraries for encoding/decoding/multiplexing for streams. However, student will also need to add more features such as GUI, multiple encoding. This project should develop an application similar to Microsoft Expression Encoder tool. There are many public tools that can be used as a baseline for implementation. Students are encouraged to start from any of these baselines and extend their functionalities throughput the project.
    Pre-requisite courses:
    
    DMET703: Video and Audio Technology

Bachelor Thesis Topics

Acoustic Environment Classification

Project Tasks:

Comparing Acoustic Environment Classification Techniques Research

Adaptive Acoustic Environment Classification Research

Ambient Computing

Project Tasks:

Using Mini-fans as an Ambient Display Research

Automatic Configuration of Combinational Logic Circuits

Project Tasks:

Digital Circuit Design using Evolutionary Algorithms for Research

Ant Colony Optimization for optimal Digital Design Research

Combinatorial Algorithms

Project Tasks:

Exploring Algorithms for Constrained String Enumeration Research

Computer Supported Collaborative Work

Project Tasks:

Multi-author Novels Research

Project Tasks:

Connected Rooms Research

Project Tasks:

Your Handwriting Your Font Research

Constraint Reasoning

Project Tasks:

Matrices in 2D and geometric reading Application

Comparing pivot rules of the Simplex algorithm Research

Project Tasks:

Staff Scheduling System Research

Project Tasks:

Portfolio selection Application

Project Tasks:

Confluence of CHR programs Research

A Java Visualization Tool for Transformed CHR Programs Research

Developing Apps for Intel Ultrabook and Win8 Platforms

Project Tasks:

Innovative SW Usage Model Application

Electrocardiogram (ECG) Processing and Analysis

Project Tasks:

Statistical and Emerging ECG Modeling Research

Project Tasks:

ECG Compression Research

Project Tasks:

ECG Classification and Analysis Research

Game with a Purpose and Crowd Sourcing

Project Tasks:

Egyptian Monument Locator Research

Egyptian Dialect to Standard Arabic Research

Overview of Image Recognition Tools for Egyptian Landmarks Application

ARTigo 2.0 - Identifying Objects Research

AreCAPTCHA - Arabic reCAPTCHA Research

GUC Software Foundation

Project Tasks:

Pseudo-code experimental interpreter Application

Courses manager Application

Major selection guide for engineering students Application

Student course evaluation Application

Student semester load simulator Application

Matching GUC majors with Alumni jobs Application

Faculty statistical tool Application

Human Activity Recognition

Project Tasks:

Gyroscopes for Human Activity Recognition Application

Hyperspectral Images

Project Tasks:

Denoising of Hyperspectral Images Application

Project Tasks:

Endmember Extraction from of Hyperspectral Images Research

Hypertext Framework

Project Tasks:

Design of the Framework and Implementing Core Services Research

Integrated meta-modeling based software engineering framework

Project Tasks:

Form-based component diagram editing feature Application

Graph-based component diagram editing feature Application

Project Tasks:

Graph-based process specification feature Application

Form-based process specification feature Application

Process execution environment Application

Project Tasks:

Behavioral model back-end translator Application