Capture Smart AI-Blur Aware Mobile Camera Control

📱 Project Overview

Aim:
To reduce motion blur in captured images by dynamically adjusting camera parameters such as shutter speed and ISO based on real-time analysis of motion, brightness, and blur. The app provides two camera modes:

• AI Camera: Analyzes each frame using OpenCV and adjusts parameters automatically.
• Normal Camera: Captures images using the default camera settings.

---

🚀 Features

• 📷 AI-Powered Camera: Real-time analysis of motion, brightness, and blur using OpenCV.
• 🔧 Dynamic Shutter Speed and ISO: Adjusts exposure based on motion/lighting conditions.
• 🔄 Camera Switch: Toggle between front and rear cameras.
• 🖼️ Overlay UI: Displays motion %, brightness, blur score, shutter speed, and ISO.
• 💡 Info Toggle: Show/hide analysis info with a button.
• 🧠 OpenCV Integration: Used for image processing and frame analysis.
• 🧪 Post-Capture Analysis (Python): Visualizes blur heatmaps using edge detection and heatmap overlay.

---

📄 View Project Report (PDF)

🧭 Navigation Flow

1. SplashActivity

   • Shows animated splash screen.
   • Navigates to the main menu after a short delay.

2. MenuActivity

   • Offers two options:
     • 📱 AI Camera (launches CameraActivity)
     • 📷 Normal Camera (launches NormalCameraActivity)

3. CameraActivity (AI Camera)

   • Uses Camera2 API and OpenCV to analyze each frame.
   • Dynamically computes and sets shutter speed and ISO.
   • Captures and saves high-quality deblurred images.

4. NormalCameraActivity

   • Basic camera capture without motion or blur handling.
   • Switch between front and back cameras.
   • Saves standard JPEG images.

---

🧪 Post-Capture Image Analysis (Python Tool)

The Python script test.py is designed to analyze images captured by the Android app and visualize the blur levels using edge detection and heatmap overlays.

🔍 What It Does

This script helps you understand how much of the image is sharp versus blurry by processing it through several image-processing stages. Here's a breakdown of the workflow:

1. Read the Input Image

   • The image is loaded using OpenCV.
   • If the image path is invalid or unreadable, it prints an error and exits.

2. Convert to Grayscale

   • The image is converted to grayscale to simplify edge detection since edges are typically detected from intensity changes.

3. Apply Canny Edge Detection

   • Detects edges in the grayscale image using two threshold values (low_thresh=50, high_thresh=150).
   • This generates a binary map showing where sharp edges exist.

4. Create a Blur Map

   • The edge map is inverted using bitwise NOT operation.
   • Areas without edges (low contrast/blurred zones) become bright in the blur map.

5. Smooth the Blur Map

   • Gaussian Blur is applied to make the visualization softer and more continuous.
   • This helps avoid harsh transitions and makes the heatmap more informative.

6. Generate Heatmap

   • A color heatmap (using COLORMAP_JET) is applied to the smooth blur map.
   • Red/blue regions indicate areas of high blur.
   • Yellow/green regions indicate sharper zones with strong edge presence.

7. Display Results

   • Uses matplotlib to display:
     • Original image (RGB)
     • Edge map from Canny
     • Heatmap showing blur distribution
   • Optionally saves the output if save_path is provided.

📈 Sample Use Case

image_file = 'bottle_without.jpg'
output_file = 'heatmap_bottle_without.png'
blur_highlight_with_edges(image_file, save_path=output_file)

🔧 Dependencies

Make sure you have the following installed:

pip install opencv-python matplotlib numpy

📌 Notes

• This script is useful for validating how well the Android camera app is reducing motion blur.
• Works best on JPEG images captured by either CameraActivity or NormalCameraActivity.
• The application is named as "Hackethon", you can directly open this folder in Android Studio.

---

🛠️ Tech Stack

• Language: Java (Android), Python (Post-processing)
• Framework: Android SDK
• Image Processing: OpenCV (Android + Python)
• Camera API: Android Camera2 API
• UI Design: XML Layouts
• Visualization: Matplotlib (Python)

---

🔧 Permissions Used

• CAMERA
• WRITE_EXTERNAL_STORAGE (for Android <= API 28)
• READ_MEDIA_IMAGES (for Android >= API 33)

Defined in AndroidManifest.xml.

---

📂 Key Files

File	Purpose
CameraActivity.java	Main AI camera logic with adaptive exposure
NormalCameraActivity.java	Standard camera logic (default Android camera)
MenuActivity.java	Entry point after splash; offers camera options
SplashActivity.java	Animated splash screen logic
test.py	Python script to generate heatmaps from captured images
AndroidManifest.xml	Declares app structure and permissions

---

✅ How to Run

Android App:

1. Clone this repository.
2. Open Hackethon directory in Android Studio.
3. Connect a physical Android device or use an emulator with camera support enabled with USB Debugging.
4. Click Run ▶️.
5. The Application should successfully install with name as Hackethon

Python Script:

1. Install dependencies:

   pip install opencv-python matplotlib numpy

2. Place captured images in the same folder as test.py.
3. Update the image_file path in the script and run:

   python test.py

   This will generate a heatmap of the blurred images.

---

---

🔗 Related Links

• 📹 Demo Video

---

📃 License

This project is for educational and research purposes.