Eren Alpar

This is an ongoing project aimed at developing a document analysis tool that extracts structured information—such as text, images, and tables—from PDF files.

The planned pipeline includes:

-> PDF to Image Conversion: Convert PDF pages into images for further processing.

-> Image Preprocessing: Enhance images using OpenCV to optimize OCR accuracy.

-> Data Extraction: Use PyTesseract or ML to extract and categorize content.

-> Structured Output: Save extracted data as a JSON file for downstream use.

-> User Interface: Provide an interactive UI built with Streamlit for uploading PDFs and displaying results.

The goal of this project is to deepen my experience with OCR technologies, image processing, and user interface development with Streamlit.

This is an experimental game project focused on procedural terrain generation using Perlin noise. While the game is not fully playable, it successfully renders a dynamic 3D terrain composed of spheres, demonstrating real-time rendering techniques and mathematical problem-solving in graphics programming.

I developed the rendering engine from scratch using C++ and OpenGL, writing custom shaders in GLSL. The goal of this project was to strengthen my understanding of computer graphics, linear algebra, and C++ object-oriented design.

Through this work, I significantly improved my skills in OpenGL, shader programming, and spatial mathematics—solving challenges related to efficient sphere rendering, chunk management, and real-time terrain generation.

This is a custom-built tokenizer and morphological analyzer developed entirely in C++ using only the standard library, with no external dependencies. The primary goal of the application is to parse English sentences both morphologically and syntactically.

The system tokenizes input text, analyzes the structure of each word, and classifies them into one of ten grammatical categories. It also includes logic for handling suffixed words, irregular verbs, and singular/plural forms.

Developed as a full-stack project, this application demonstrates my deep understanding of C++ programming, English language rules, and natural language processing fundamentals. For more detailed information, please refer to the report.pdf available on GitHub.

This is my second hyper-casual Android game, developed and published on the Google Play Store. As with my previous game projects, I independently handled the full development cycle—including game design, coding, 3D asset creation, and UI/UX design.

The game was built using the Unity engine with C#, and all 3D assets were created in Blender. Textures and visual elements were designed using Gimp. The source code is version-controlled in a local Git repository and can be shared upon request.

This project further solidified my skills in mobile game development, rapid prototyping, and delivering a polished product to a public audience.

This project is a command-line calculator capable of parsing and evaluating complex algebraic expressions. It begins by sanitizing the input (e.g. removing spaces, validating characters), then processes the expression using a custom parsing algorithm that respects mathematical order of operations and properly handles nested parentheses.

The calculator supports operations such as addition, subtraction, multiplication, division, exponentiation, and modulus. It evaluates expressions from the innermost parentheses outward, ensuring correct left-to-right precedence within each level.

Example input: (((5 + 6) * 45) ^ 3 / 77 + 17 - (4.5 * -4) + 174) / (5^4 % 7) * (8^5)

This is an image processing application that detects and classifies six different Turkish Lira banknotes using the OpenCV library. The system determines the value of a banknote based on its visual features and can recognize it regardless of its orientation by analyzing both the front and back sides.

The project demonstrates my proficiency in OpenCV and core image processing techniques such as contour detection, edge filtering, template matching, and rotation handling.

Enadream.com is a personal portfolio website designed to showcase my projects, share updates, and provide a central platform for communication. I developed the entire site from scratch using HTML, CSS, JavaScript, and PHP, with deployment on a Linux-based Amazon AWS server.

This project strengthened my skills in full-stack web development, server configuration, and content management. It reflects my ability to design and maintain a responsive, functional, and visually engaging website.

This was my first completed and publicly released game, available on the Google Play Store. I independently managed the entire development process—from concept and design to coding and publishing. Using Unity and C#, I implemented all game mechanics and systems. I created 3D assets in Blender and designed custom textures in Adobe Photoshop.

This project gave me hands-on experience with full-stack game development, including gameplay design, asset creation, UI development, and optimization for mobile platforms.