Skip to content

Mak-3/Car-Dirtiness-and-Damage-detection

Repository files navigation

Project Name

CAR DIRTINESS AND DAMAGE DETECTION WITH SPECIFICATION

Purpose

Car maintenance is really a hectic task when it comes to waiting for longer duration hence we can automate this task by recommending users based on their car condition

Requirements

  • Tensorflow
  • Flask
  • PIL
  • matplotlib
  • requests
  • scipy
  • torch
  • pandas

How to run this code

git clone https://github.com/Mak-3/Car-Dirtiness-and-Damage-detection
cd Car-Dirtiness-and-Damage-detection
python flask_app.py
open in localhost

output

Screenshot (14) Screenshot (24) Screenshot (28)

Types of output

YOLOv5 will return image if damaged with bounding boxes namely of 4 possibilities

  • Scratch
  • Glass broken
  • Deformation
  • Broken

VGG16 will return output in String format in the cmd as either "messy" or "clean"