Overview

Inspectra is my Final Year Project — a mobile application that automates vehicle damage assessment using computer vision. A user photographs a damaged vehicle and receives a detailed damage report with severity classification and local repair cost estimates within 60 seconds, eliminating the need for a manual inspection.

Highlights

• ▸Built mobile application automating vehicle damage assessment in under 60 seconds using YOLOv8; outputs severity-classified damage reports with cost estimates based on local repair pricing.
• ▸Designed microservices backend with Node.js, FastAPI, PostgreSQL, and Supabase for async processing and real-time alerts.
• ▸Engineered 8-stage ML pipeline: preprocessing → YOLOv8-seg → part detection → severity classification → report generation.