🖼️ Image Recognition using AWS Rekognition

Frontend: React (Vercel Hosted)
Backend: AWS EC2 + API Gateway
Image Processing: AWS Rekognition
Storage: Amazon S3 (with lifecycle policy)

---

📌 Overview

A cloud-based image recognition system leveraging AWS Rekognition to detect and analyze objects in images.

Key Points:

• Amazon S3 temporarily stores uploaded images.
• Lifecycle Policies delete images automatically after 1 day.
• Vercel hosts the React frontend for a smooth user experience.
• Ensures cost-efficiency, privacy, and scalability.

---

🏗️ Cloud Architecture

          🖥️ User
            │
            ▼
   🎨 Frontend (React on Vercel)
            │
            ▼
    🌐 API Gateway (REST API)
            │
            ▼
  ⚙️ Backend (AWS EC2 Server)
            │
            ▼
 📂 S3 Bucket (Auto-delete in 1 Day)
            │
            ▼
🔍 AWS Rekognition (Image Analysis)

---

##🚀 Features

• Upload an image for real-time recognition
• Detects objects, scenes, and faces
• Auto-deletes images after 1 day (S3 Lifecycle Policy)
• Secure communication via IAM Roles & API Gateway
• Fully deployed backend on AWS EC2
• Responsive Vercel-hosted frontend

---

##🛠️ AWS Services & Responsibilities

Service           Purpose
----------------  ---------------------------------------------------
Amazon EC2        Host backend API server
API Gateway       Route API requests securely
Amazon S3         Temporary storage for uploaded images
Lifecycle Policy  Auto-delete files after 1 day
AWS Rekognition   Image analysis for objects, scenes, and faces
IAM               Secure AWS resource access
Vercel            Host React frontend

---

##💼 My Role as Cloud Engineer

• Architected entire AWS infrastructure
• Deployed backend API on Amazon EC2
• Integrated API Gateway for secure requests
• Configured S3 Lifecycle Policy for auto-deletion
• Implemented recognition workflow using AWS Rekognition
• Applied least-privilege IAM roles
• Hosted frontend on Vercel
• Delivered cost-efficient, automated image processing

---

##🌐 Live Deployment

Backend API https: https://.execute-api..amazonaws.com/prod

Frontend (Vercel): https://image-recognition-using-aws-rekogni.vercel.app/

---

##🧪 Local Development

#1️⃣ Clone the repo

git clone https://github.com/pardeep1916P/image-recognition-using-aws-rekognition.git
cd image-recognition-using-aws-rekognition

#2️⃣ Install frontend dependencies

npm install

#3️⃣ Install backend dependencies

cd ../server
npm install

#4️⃣ Start backend locally

npm start

---

##📂 Project Structure

client/
 ├── build/
 │    ├── static/                 # Compiled frontend assets
 │    ├── asset-manifest.json
 │    ├── index.html
 │    └── logo.png
 ├── public/
 │    ├── index.html               # Public HTML entry point
 │    └── logo.png
 ├── src/
 │    ├── components/              # React components
 │    │    └── App.tsx
 │    ├── index.css
 │    ├── index.tsx
 │    └── utils.ts                  # Helper functions
 ├── package-lock.json
 ├── package.json
 └── tsconfig.json
server/
 ├── index.js                       # Backend entry point
 ├── package-lock.json
 └── package.json

---

##✅ Impact

• Created a secure, scalable AWS-based image recognition platform
• Reduced storage costs via automatic deletion
• Ensured high availability with EC2 deployment
• Improved data privacy with API Gateway + IAM
• Delivered a modern, fast, and responsive UI via Vercel