🔫 Weapon Detection System using CNN and Flask

📌 Objective

The primary objective of this project is to develop a Weapon Detection System that can accurately identify whether an image contains a weapon or not. The system utilizes a Convolutional Neural Network (CNN) model for classification and is deployed using a Flask web application, providing an easy-to-use interface for end-users to upload images and receive detection results in real-time.

---

🌟 Features

• ✅ Detects presence or absence of weapons in uploaded images
• 🧠 CNN-based custom model for high accuracy
• 💻 Flask-powered lightweight web application
• 📤 Image upload interface
• ⚡ Real-time prediction
• 📊 Clean and organized codebase for easy understanding
• 🧪 Easily extendable for video surveillance integration

---

🧠 Model Architecture

• Model Type: Convolutional Neural Network (CNN)
• Input Shape: Preprocessed image (e.g., 128x128 or 224x224 pixels)
• Layers:
  • Convolutional Layers
  • Max Pooling Layers
  • Flatten Layer
  • Dense Layers
  • Dropout Layer
  • Output Layer with Softmax/Sigmoid Activation
• Output Classes: Weapon, No Weapon
• Loss Function: Binary Crossentropy
• Optimizer: Adam
• Metrics: Accuracy, Precision, Recall

---

🛠️ Implementation

1. Data Collection & Preprocessing

• Gathered a dataset of images containing weapons and non-weapons.
• Preprocessed images: resizing, normalization, and augmentation.
• Split into training, testing and validation sets.

2. Model Training

• Built and trained the CNN using TensorFlow/Keras.
• Evaluated performance using accuracy, precision, and recall.
• Saved the trained model using model.save('weapon_detector.h5').

3. Flask Integration

• Created a Flask application to serve the model.
• Users can upload an image via a simple UI.
• The server loads the saved model, processes the image, and returns prediction results.

4. UI/UX

• HTML form with file input for image upload.
• Display area to show image and classification result.
• Error handling for invalid files or upload issues.

---

🚀 How to Run the Project

1. Clone the Repository

git clone https://github.com/Jerry-britto/Weapon-Detection.git
cd Weapon-Detection/app

2. Install Dependencies

pip install -r requirements.txt

3. Run the Flask App

python app.py

4. Access the Web Interface

Open your browser and go to: http://127.0.0.1:5000

---

🧪 Sample Usage

1. Open the web interface.
2. Upload an image.
3. Click the "Detect" button.
4. Get prediction result: “Weapon Detected” or “No Weapon Detected”.

---

🛡️ Use Cases

• Surveillance and security monitoring systems
• Crime prevention tools
• Restricted area monitoring (schools, airports, etc.)
• Law enforcement decision-support systems

---

📈 Future Enhancements

• Integrate with live video feed for real-time detection
• Add bounding box localization for weapon in the image
• Implement confidence score display
• Deploy to cloud (AWS/GCP/Azure)
• Mobile app integration using APIs