Wow, a whole page about me!

Work-related Projects
Remote Student Monitor Framework for Securing Exams

Main and only Developer for building a remote exam proctoring platform using React/MUI/Typescript and Google Functions. The platform continuously monitors students and detects cheating behavior, triggering video recordings for manual verification by the proctor.
sWaiver – Signed Forms

Online management system for handling signed forms.
- Used Ionic/Angular for the front end and Google Functions for the back end
- User-friendly and effectively handles signed forms
- Involved clients/consumers interaction and feedback incorporation
Graphical User Interface for Hellenic Autonomous Vehicle

This subcontract concerns the HAV Graphical User Interface (HAV-GUI), an interface developed for car touchscreens. I constructed the HAV-GUI using NVIDIA’s DriverWorks SDK, and it consists of three key components: a tile/grid display that allows users to view all available information screens, button/tile elements that enable users to activate specific functions, and a layout engine that arranges these elements on the screen.
- I communicate with multiple academic institutions and companies to create and implement the required specifications.
- Create custom GUI elements using C++ and the NVIDIA DriverWorks SDK.
Ship Emissions Monitoring System
Primary backend developer for a system to monitor ship emissions. Collect data from multiple IOT sensors and store them in a series-type database (InfluxDb). Then, the data will be extracted, and the CO2 emissions and fuel consumption will be calculated.
Research Projects
Unsupervised Word Spotting in Historical Handwritten Document Images Using Document-Oriented Local Features

A new method that permits effective word spotting in handwritten documents is developed that relies upon document-oriented local features, which take into account information around representative key points as well a matching process that incorporates spatial context in a local proximity search without using any training data (paper)
- I proposed a novel idea of search symbols/words based on image representation in extensive image collections.
- Create custom data structures and algorithms to facilitate many-to-many matching between multiple points (X, Y) (with an associate vector/set of weights) in a large set of images.
- The front end was developed using Angular, and the back end was developed using C#/.NET.
Computer-aided diagnosis of mammographic masses

Supervised and contributed to developing a recommendation system that comprises two stages: A margin-specific supervised Content-Based Information Retrieval (CBIR) stage that retrieves images from reference cases along with a decision stage that is based on the retrieved items. (paper)
- It was developed using Python and the following libraries: numpy, pandas, scikit-learn, and OpenCV.
- Needed to communicate with multiple academic institutions and people to gather/create the necessary data.
- I supervised the development/research of the system and contributed specifically to the development/research of the recommendation system.
- The recommendation system is based on a decision involving multiple SVMs modelling different types of cancer based on the distance from each SVM hyperplane.