⚙️ EyeDentifyAPI: Ktor Backend (API Orchestration)

This repository contains the backend server for the EyeDentify mobile application, built using the Ktor framework in Kotlin. Its primary function is to securely manage user data, handle file uploads, and orchestrate calls to multiple specialized AI services.

🔗 Related Repository

• EyeDentify (Android Client)

🚀 Key Technologies & Services

Technology	Purpose
Ktor Framework	Asynchronous server engine (Netty) for routing and handling multipart requests.
PostgreSQL / JDBC	Data persistence for users (BCrypt hashing) and history storage (BLOB/BYTEA).
BCrypt	Used for secure, one-way hashing of user passwords during registration and login.
Google Vision API	Primary general object identification service.
OpenAI (GPT-4o)	Advanced multi-modal identification and structured JSON response generation.
Pl@ntNet API	Specialized identification service for plants, flowers, and trees.

💻 Running the Backend

Prerequisites

• JVM (Java Development Kit) installed.
• A running PostgreSQL instance.
• API Keys for Google Vision, OpenAI, and Pl@ntNet.

Setup and Configuration

1. Clone the repository:

   git clone https://github.com/JPR420/EyeDentifyAPI.git
   cd EyeDentifyAPI

2. Database Setup: Ensure your PostgreSQL instance is running. The server attempts to create the required users and capture_results tables on startup (code is commented out in main for safety, may need manual execution).

3. Credential Management: You must configure your database and API credentials in the DatabaseConfig.kt file (or a secure environment variable system):

   • jdbcUrl, username, password (for PostgreSQL)
   • googleKey, openAiKey, plantNetKey

4. Start the Server: Run the main function in your IDE, or use Gradle:

   ./gradlew run
   # Server will start on http://localhost:8080

🗺️ API Endpoints

Endpoint	Method	Role	Body/Params
/register	POST	Creates a new user with BCrypt hashed password.	email, password (Form Parameters)
/login	POST	Authenticates user and returns id and tier.	email, password (Form Parameters)
/identify	POST	General object recognition via Google Vision.	Multipart (Image File)
/identify_openai	POST	Detailed object recognition via GPT-4o-mini.	Multipart (Image File)
/identify_plantnet	POST	Plant/Nature identification via Pl@ntNet.	Multipart (Image File, optional organs)
/saveResult	POST	Saves captured data and image to user history.	Multipart (Image File + Form data: user_id, object_name, etc.)
/getUserHistory	POST	Retrieves history for a user, returning images as Base64.	user_id (Form Parameter)