Sprint 1 Plan -- Smart Parking System

Sprint 1 focuses on building the core foundation of the Smart Parking System over two weeks. The primary objectives to complete are the Automated License Plate Capture System, AI Data Processing & OCR, and Database Storage & Schema.

For the license plate capture system, the team will integrate cameras with parking spot sensors, implement a YOLO model for plate detection, and build an image preprocessing pipeline covering brightness adjustment and noise reduction. Detection accuracy will be tested across different lighting conditions, and edge cases such as dirty or obscured plates will be handled.

On the AI and OCR side, the team will train and fine-tune an OCR model for extracting license plate text, build an API endpoint that receives camera images and returns plate text, implement format validation, set up error logging for failed detection, and create a fallback mechanism for cases where the model's confidence is low.

For the database, the team will design a PostgreSQL schema for parking sessions, create tables for vehicles, spots, floors, and timestamps, enforce unique constraints to prevent duplicate spot assignments, set up indexing for fast license plate lookups, and write migration scripts for initial setup.

In addition to these new objectives, two features from prior work will continue through this sprint. The manual check-in feature will see the UI form completed with input validation, error messaging, and backend API integration. User authentication will progress with basic registration and logic, session management, user database tables, and password hashing.

To track and manage all of this work on GitHub, the sprint will be organized as a milestone with a progress bar monitoring open versus closed issues across the entire sprint. A project board will divide all tasks into To Do, In Progress, and Done columns so the team has a clear view of momentum at any point. All issues will be grouped by epic -- ALPC, OCR, DB, AUTH, and MCI -- with individual task checklists inside each issue, assigned issue numbers for easy cross-referencing, and labels for each category, including camera, AI/ML, backend, database, security, and frontend. Epic tags will be applied to every issue so that no task exists in isolation from its parent feature, making it easy to filter and track progress at both the task and feature level.

By the end of Sprint 1, the system should have a functioning detection pipeline, a structured database, and solid groundwork for both authentication and manual check-in handing into Sprint 2.