Abhijeet Prajapati

This project features a pathfinding visualizer utilizing a 2D array and displays the result on a pygame canvas where users can draw terrain, set a starting point, and define a goal. The algorithm (Manhattan Distance) then calculates and displays the shortest path through the user-created terrain, providing an interactive way to understand pathfinding algorithms.

Simple terrain generation program created in JavaScript with the p5.js library. It utilizes Perlin noise to create landscapes, distinguishing water, land, grass, trees, and mountains with varying colors. The gradient in colors indicates height changes within each terrain type. Users can adjust the seed, zoom in/out, and save the terrain as a .png file.

Our project focused on creating a deep learning model using the Keras library to classify handwritten digits from the MNIST dataset. Our model achieved an accuracy of 98% during training. Users can easily interact and test the model with the GUI, drawing digits for accurate recognition by the trained model.

Our semester-long course project resulted in the development of a fast and responsive website for the Niagara-on-the-Lake Museum. This website serves as an immersive digital showcase for all museum exhibits, providing visitors with accurate information about each display. Key features include a scheduling system that enables visitors to plan their museum visit conveniently. Additionally, an admin page allows museum staff to update exhibit information, which is stored in Supabase, seamlessly. Notably, our project received recognition from our professor, who proposed its adoption by the actual museum.

This project is an AI-powered travel planning app built with React.JS and Squid AI. It helps users plan trips by suggesting destinations, places to stay, things to do, and travel routes based on their preferences—like budget and travel style. The app also adjusts plans on the fly, giving real-time updates and alternatives to make sure the trip goes smoothly. The goal is to make planning trips super easy and fun, with an intuitive interface that takes the hassle out of organizing everything.

Developed a fully featured chess game in Python using Pygame. The game can be played against either an AI opponent or another human player. It ensures all moves are valid and leverages the Minimax algorithm with alpha-beta pruning to enable the AI to make the best possible moves within the user's selected depth.

Developed a Python-based desktop application that automates screen monitoring through intelligent screenshot capturing and button detection. The tool allows users to set custom screenshot intervals, capture specific screen regions, and track on-screen elements with precision. By leveraging computer vision techniques and a modern, dark-themed Tkinter interface, I created a versatile automation tool that can be used for various tasks like game farming, website monitoring, or custom workflow automation. Built using libraries like OpenCV, PyAutoGUI, and Python's threading module.

The program is a user-friendly GUI application designed to simplify the process of downloading YouTube videos. Built using Python, it integrates YT-DLP and FFmpeg to provide a robust and flexible downloading experience. Users can easily select from a variety of options, such as downloading video or audio-only formats, choosing the desired quality, and specifying the save location. The intuitive interface ensures that even non-technical users can effortlessly navigate and customize their downloads, making it a versatile tool for anyone looking to save YouTube content locally.