Maatrika P's Projects
Revolutionary Chatbot Greeting System: Seamlessly Welcoming Users through Face Detection Technology
This project aims to develop an automated system for segmenting containers based on liquid levels. By detecting variations and errors in the levels, the system predicts if a bottle is accepted or rejected, ensuring accurate quantification and quality control on a conveyer belt.
Explore data preprocessing and machine learning with this Jupyter Notebook, covering essential techniques and algorithms
This repository contains code for leveraging Flan-T5 for initial exploration and fine-tuning the T5-Small model to generate concise and informative summaries from text data.
Developed a solution to predict the value of Kp index, calculated from geomagnetic field data across 13 locations on Earth & data from DSCOVR satellite. Used ARIMA, to make these predictions based on Level 1 time series data provided by NASA. Kp index prediction gives us valuable insights into current Geomagnetic Weather.
This project aims to improve image registration accuracy by integrating feature extraction and machine learning techniques. It involves a multi-step process, including preprocessing, feature detection, robust estimation, and perspective transformation. Machine learning classifier is trained using extracted features.
This project demonstrates the development of an intrusion detection system using machine learning techniques such as LSTM and CNN to safeguard networks against malicious activities and cyber threats.
By employing the UNet-VGG architecture, we improve terrain understanding, enabling safer and more efficient lunar rover navigation, paving the way for advanced lunar exploration.
Developer leveraging technology to solve real-world problems and create innovative solutions.
Find my recent project works on my portfolio. If you like what you see, star it!
Revolutionize education with a Neuroadaptive Recommender System. Personalize learning using OpenBCI data & learning analytics. Enhance outcomes by bridging gaps, adapting in real-time. Empower learners, educators, institutions. Engaging, data-driven EdTech solution
This code implements road segmentation using an FCN model. It loads a dataset, defines the model architecture with convolutional and upsampling layers, trains the model, evaluates its performance, and displays example predictions. The goal is to accurately segment roads in images.
Unlock the potential of autonomous vehicles with advanced object detection and segmentation. Harness the BDD100K dataset to enhance road safety and transportation efficiency
Ship Detection in SAR Images: A repository for analyzing, training, and testing ship detection models using Synthetic Aperture Radar (SAR) images.
Utilized Social Network Analysis to explore Facebook data, identifying key individuals, and visualizing network patterns to gain insights into online social connections and interactions.
This project integrates Bing API for geocoding, OpenWeatherMap API for weather data, and leverages GitHub Copilot for AI-generated code. Fetch weather for cities or user location, display temperature/humidity with color-coded formatting. Streamline development with AI assistance.
Web Tech Lab: Exploring JS, CSS, HTML, PHP, JSP projects for hands-on learning and innovation.