A comprehensive study on detecting DDoS attacks using machine learning techniques in the OMNeT++ simulation environment. The project analyzes network traffic features including packet size, ports, and transmission frequency. Using various ML models (Logistic Regression, Decision Trees, MLP, Random Forest), achieved 94% accuracy and 0.927 ROC-AUC score with Random Forest, demonstrating effective identification of network anomalies.

Implementation of a road line detection system using computer vision techniques. The project processes video input to identify and highlight lane lines, essential for autonomous driving applications.

A sophisticated computer vision pipeline for robust lane detection in challenging driving conditions. Features camera calibration, perspective transform, color/gradient thresholding, and polynomial fitting for accurate lane boundary detection.

This ROS2 package simulates real-time GPS and IMU data streams by publishing NavSatFix, Imu, and PoseWithCovarianceStamped messages. Perfect for testing autonomous systems without hardware.

Understanding commonly used hardware for self-driving cars, identifying the main components of the self-driving software stack, programming vehicle modeling and control.

A comprehensive analysis of machine learning models for heart attack prediction using various analytical techniques including EDA, correlation analysis, and predictive modeling.

Implementation of three critical phases of autonomous driving planning: Short-Term, Immediate-Term, and Behavior Planning, focusing on obstacle avoidance and traffic rules.

Application of Convolutional Neural Networks for flower recognition using MobileNetV2 model, utilizing a dataset of 4242 images across five flower types.

Classification of urban noise using various machine learning models including DNN, CNN, LSTM, and Random Forest using the UrbanSound8K dataset.

Developed a real-time image stitching system optimized for low to mid-range smartphones that processes video frames during the recording process to create seamless panoramic images. The project leverages efficient feature detection, matching, and homography estimation techniques for accurate stitching while maintaining smooth performance on mobile devices with limited computational resources.

Implementation of a 2D particle filter to solve the "kidnapped car" problem for vehicle localization. The algorithm predicts positions, matches sensor data to landmarks, updates weights with Gaussian distributions, and resamples particles for accurate vehicle tracking.

An innovative project utilizing artificial intelligence to analyze ultrasound images for early detection of breast cancer. The system aims to improve diagnostic accuracy and reduce false positives through advanced machine learning algorithms and image processing techniques.