New Features

Author: maheshmm7

README.md

@@ -1,31 +1,50 @@
-# DrowsiGaurd (Drowsiness Detection System)
+# DrowsiGuard (Drowsiness Detection System)
 
-This project implements a **Drowsiness Detection System** using **Computer Vision** and **Deep Learning** techniques to monitor and detect signs of drowsiness in drivers. The system tracks the driver's eye state and gaze direction using the webcam, and it raises an alert if drowsiness is detected, ensuring safer driving conditions.
+> **Note:** This README is for the `logfeature` branch which includes additional terminal output for facial landmarks detection.
 
-## Features
+This project implements a **Drowsiness Detection System** using **Computer Vision** and **Deep Learning** techniques to monitor and detect signs of drowsiness in drivers. The system uses webcam input to track eye states, gaze direction, and facial landmarks, raising alerts when signs of drowsiness are detected.
+
+## New Features in This Branch
 
-- **Real-time Drowsiness Detection:** Tracks the eye state to detect when the driver is drowsy or blinking excessively.
-- **Gaze Direction Tracking:** Monitors the direction of the gaze to determine if the driver is distracted.
-- **Head Pose Estimation:** Analyzes the orientation of the driver's head to detect abnormal head movements or drowsiness.
-- **Alert System:** Plays an alarm sound when drowsiness or distraction is detected.
-- **Face and Eye Detection:** Utilizes Haar cascades for detecting faces and eyes.
+- **Real-time Landmark Printing:** Continuously prints coordinates in terminal for:
+  - Left eye landmarks
+  - Right eye landmarks
+  - Complete facial landmarks (x, y, z coordinates)
+- All other features from the main branch remain unchanged
+
+## Features
 
+- **Real-time Eye State Monitoring:** Uses a CNN model to classify eyes as open or closed
+- **Drowsiness Score System:** Maintains a drowsiness score based on eye state patterns
+- **Gaze Direction Tracking:** Uses MediaPipe face mesh to monitor the driver's gaze direction
+- **Alert System:** Plays an alarm sound when prolonged drowsiness is detected
+- **Event Logging:** Records drowsiness events with timestamps
+- **Image Capture:** Saves images when drowsy or sleepy states are detected
+- **Visual Indicators:** Dynamic visual overlays indicating different drowsiness levels
 
 ## Prerequisites
-- **Python Version**: Python 3.7 - 3.9 (Python < 3.10 required)
+
+- Python 3.7 - 3.9 (Python < 3.10 required)
 - Webcam for real-time monitoring
-- The following dependencies, listed in requirements.txt file
+- Required libraries (listed in requirements.txt)
 
 ## Setup and Installation
 
-### 1. Using Virtual Environment (venv)
+### 1. Clone the Repository
+```bash
+git clone <repository-url>
+cd drowsiguard
+git checkout feature/landmark-printing
+```
 
-#### 1.1 Create Virtual Environment
+### 2. Using Virtual Environment (venv)
+
+#### 2.1 Create Virtual Environment
 ```bash
 python -m venv drowsiness_env
 ```
 
-#### 1.2 Activate Virtual Environment
+#### 2.2 Activate Virtual Environment
 - **Windows**: 
   ```bash
   drowsiness_env\Scripts\activate
@@ -35,136 +54,117 @@ python -m venv drowsiness_env
   source drowsiness_env/bin/activate
   ```
 
-#### 1.3 Install Dependencies
+#### 2.3 Install Dependencies
 ```bash
+pip install --upgrade pip
 pip install -r requirements.txt
-pip install protobuf==3.20.0
+pip install protobuf==3.20.0  # Required for MediaPipe compatibility
 ```
 
-### 2. Using Conda Environment
+### 3. Using Conda Environment
 
-#### 2.1 Create Conda Environment
+#### 3.1 Create Conda Environment
 ```bash
 conda create --name drowsiness_env python=3.9 -y
 ```
 
-#### 2.2 Activate Conda Environment
+#### 3.2 Activate Conda Environment
 ```bash
 conda activate drowsiness_env
 ```
 
-#### 2.3 Install Dependencies
+#### 3.3 Install Dependencies
 ```bash
+pip install --upgrade pip
 pip install -r requirements.txt
-pip install protobuf==3.20.0
+pip install protobuf==3.20.0  # Required for MediaPipe compatibility
 ```
 
-### 3. Verify Setup
-
-Check installed packages:
+### 4. Verify Installation
 ```bash
-pip list
-```
-
-## Recommendations
-
-- **Conda is recommended** for managing complex dependencies, especially for machine learning projects with potential GPU support.
-- Always ensure you're in the correct virtual environment before running the project.
-- Keep your `requirements.txt` updated with all necessary dependencies.
-
-## Troubleshooting
-
-- If you encounter any installation issues, ensure you have the latest version of pip:
-  ```bash
-  pip install --upgrade pip
-  ```
-- Check Python version compatibility (Python 3.7 - 3.9 recommended)
-- Verify all system dependencies are installed
-
-Required libraries include:
-```text
-opencv-python
-mediapipe
-numpy
-keras
-pygame
+pip list  # Check installed packages
 ```
 
-### 4. Model and Haar Cascade Files
-Ensure the following files are available in the project directory:
+### 5. Required Files
 
-Pre-trained model for eye state classification:
-- `cnnfinal.h5`
+The system needs the following files in the project directory:
 
-Haar cascade files for face and eye detection:
-- `haarcascade_frontalface_alt.xml`
-- `haarcascade_lefteye_2splits.xml`
-- `haarcascade_righteye_2splits.xml`
+1. **Model File:**
+   - `cnnfinal.h5` (Pre-trained CNN model for eye state classification)
 
-These files can be downloaded from the official OpenCV repository or extracted from your local OpenCV installation.
+2. **Haar Cascade Files:**
+   - OpenCV's pre-trained Haar cascade files (automatically accessed via cv2.data.haarcascades)
 
-### 5. Alarm Sound
-Place an `alarm.wav` file in the project directory. This sound will be played when the system detects drowsiness or distraction.
+3. **Audio File:**
+   - `alarm.wav` (Sound file for drowsiness alerts)
 
 ## Usage
 
-### 1. Running the Script
-To start the drowsiness detection system, run:
+### Running the System
+
+1. Start the program:
 ```bash
-python test.py
+python drowdlog.py
 ```
-This will open the webcam feed and start tracking the driver's eye state, gaze direction, and head pose in real-time.
-
-### 2. How It Works
-- **Eye State Detection**: The system uses a Convolutional Neural Network (CNN) to classify whether the driver's eyes are open or closed.
-- **Face and Eye Detection**: Haar cascades are used to detect the driver's face and eyes. The system tracks the eye region to monitor blinking and drowsiness.
-- **Gaze Direction Tracking**: Using MediaPipe, the system estimates the gaze direction to detect whether the driver is distracted or looking away from the road.
-- **Head Pose Estimation**: The system tracks the head's orientation to determine if the driver is nodding or tilting their head abnormally.
-- **Alert System**: If drowsiness is detected (closed eyes for a certain duration or excessive blinking), an alarm sound is played.
-
-### 3. Exiting the Program
-To exit the program, press `q` in the terminal or stop the script execution.
-
 
+2. The system will:
+   - Access your webcam
+   - Start monitoring eye states
+   - Track gaze direction
+   - Display real-time information on screen
+   - Print landmark coordinates in the terminal:
+     ```
+     Face coordinates: x=224, y=167, w=232, h=232
+     Right Eye coordinates: x=271, y=223, w=54, h=54
+     1/1 [==============================] - 0s 25ms/step
+     Left Eye coordinates: x=273, y=224, w=53, h=53
+     1/1 [==============================] - 0s 32ms/step
+     ```
+
+## Terminal Output
+
+When running the system, you'll see continuous output in the terminal showing:
+- Face landmark coordinates (x, y, z)
+- Left and right eye bounding box coordinates
+- Detection status and changes
+
+This feature is useful for:
+- Debugging landmark detection
+- Validating eye tracking accuracy
+- Development and testing of new features
 
 ## Troubleshooting
 
-- **Webcam not detected**: Ensure that your webcam is properly connected and accessible by other applications. Try restarting your machine if the issue persists.
-- **Model loading errors**: If you get an error related to loading the model (`cnnfinal.h5`), make sure the file is placed in the correct directory and the path is specified correctly in the script.
+Common issues and solutions:
 
-## Reporting Issues and Contributing
+1. **Webcam Not Detected:**
+   - Check webcam connection
+   - Verify webcam permissions
+   - Try different video capture device indices
 
-### Reporting Errors or Bugs
+2. **Model Loading Errors:**
+   - Ensure `cnnfinal.h5` is in the correct directory
+   - Verify Keras version compatibility
 
-1. **Check Existing Issues**
-   - Before reporting, search existing GitHub issues to avoid duplicates.
+3. **MediaPipe Errors:**
+   - Check Python version compatibility
+   - Verify MediaPipe installation
 
-2. **Create a Detailed Bug Report**
-   - Use GitHub Issues
-   - Include:
-     - Detailed description of the error
-     - Steps to reproduce
-     - Your environment details (OS, Python version, etc.)
-     - Full error traceback
-
-
-- Discuss in project discussions/comments
+4. **Terminal Output Issues:**
+   - If landmark coordinates aren't printing, check if MediaPipe face mesh initialization is successful
+   - Verify console/terminal has sufficient scroll buffer for continuous output
 
 ## Contributing
 
-We welcome contributions to this project! If you'd like to contribute, please follow these steps:
-
-1. Fork the repository.
-2. Create a new branch for your changes.
-3. Commit your changes.
-4. Push your changes and create a pull request.
+Contributions are welcome! Please feel free to submit pull requests or report issues.
 
 ## License
 
-This project is licensed under the MIT License - see the LICENSE file for details.
+This project is licensed under the MIT License.
 
 ## Acknowledgments
 
-- **OpenCV**: For providing the face and eye detection algorithms.
-- **MediaPipe**: For real-time gaze and head pose tracking.
-- **Keras**: For providing the deep learning framework used to train the eye state classification model.
+- OpenCV for computer vision capabilities
+- MediaPipe for face mesh detection
+- Keras for deep learning functionality
+- pygame for audio alert system