About the Project

For my team's submission for the TMMC Innovation Challenge, we developed a program that utilizes computer vision and machine learning to identify and precisely locate any holes in automotive components. Over 300 images were collected, processed, and labeled manually to train a convolutional neural network model.

Built With

• Python
• OpenCV
• YOLOv8

Getting Started

Prerequisites

None!

Installing & Executing Program

It is recommended to run the program on Google Colab for the most consistent results.

• Upload all YOLOv8 files to Google Drive

• Mount Google Drive in Colab by running the following code:

  from google.colab import drive
  drive.mount('content/drive')
  

• Run all code segments of the Colab notebook

Authors

Charles Eret, Brandon Kong, Sean Zhang