🔍🧠 AWS Rekognition Image Analyzer with Google Colab

This project demonstrates how to securely authenticate with AWS IAM, analyze an image using AWS Rekognition, and extract meaningful labels such as objects and categories .

---

📁 Project Overview

• 🔐 Securely input AWS credentials using getpass (hidden input).
• ✅ Validate IAM session with AWS STS to confirm authorized access.
• 🖼️ Load an image from a direct URL.
• 🔄 Convert the image into byte format for Rekognition.
• 🧠 Analyze the image using AWS Rekognition's Label Detection API.
• 📝 Extract and print the top labels with confidence scores.

---

✅ Requirements

• AWS account with Rekognition enabled.
• IAM user with permissions to use Rekognition and STS.
• Your Access Key ID and Secret Access Key.
• A direct URL to an image (must return an image file, not a webpage).

---

🔧 Installation

Install boto3 in your Colab notebook:

%pip install boto3

---

⚠️ Important Notes

• The image URL must be direct to an image file, not a webpage or redirect.
• Rekognition supports images via:
  • Raw bytes (as shown).
  • Images stored in S3 (using 'S3Object': {Bucket, Name}).
• Avoid hardcoding AWS credentials; always use secure input methods.
• Use least privilege IAM policies for security.
• AWS Rekognition costs apply beyond free tier limits. Monitor usage accordingly.

---

📚 References & Resources

• AWS Rekognition Documentation
• Boto3 Rekognition Client
• AWS IAM Best Practices
• Google Colab

---