Cloud - Photo Album Setup

This guide explains how to set up a cloud-based photo album using AWS services including OpenSearch, Lambda, S3, API Gateway, and Lex.

Visit the following links for more information on setting up and managing the photo album project:

• For the S3 bucket setup, check out the repository here.
• For details on the Lambda functions setup, visit the repository here.
• If you need more detailed steps and configurations, you can find them here.

Quick overview of the setup

1. Set Up OpenSearch

Create an OpenSearch Domain: Name it photos.

2. Upload and Index Photos

• Create S3 Bucket: Name it b2-imgs.
• Lambda Function for Indexing (LF1): Name it index-photos.
  • Set Up Trigger: Trigger LF1 when a photo is added to b2-imgs.
  • Function Tasks: When an image is uploaded:
    • Use AWS Rekognition to detect image labels.
    • Retrieve metadata from the image using S3 SDK’s headObject method.
    • Create and store a JSON array of labels in OpenSearch.
• Permissions: Ensure LF1 and OpenSearch can interact by sharing their IAM role ARNs.

3. Search Functionality

• Lambda Function for Searching (LF2): Name it search-photos.
• Set Up a Lex Bot: Name it search-bot for handling search queries.
  • Intents: Create intents to capture one or two keywords from user queries for searching photos.
• Permissions: Same as step 2, ensure LF2 and OpenSearch can interact.

4. API Gateway

• Create a REST API: Import settings from a Swagger or Open API file.
• Configure Methods:
  • PUT: For uploading images to S3.
  • GET: For retrieving data from OpenSearch to display results.
• Deploy API: Create a new stage and generate an SDK for frontend integration.

5. Frontend Setup

• Web Pages:
  • Index.html: Main landing page.
  • Gallery.html: Search gallery using keywords.
  • Upload.html: Upload new images to the gallery.
• Deploy on S3: Upload these files to an S3 bucket named photo-gallery-fe and set it to allow public access.

Technologies used

aws lambda s3 api-gateway rekognition lex opensearch