/dc-logo-yolov8

Computer vision using YOLOV8 for logo recognition and identification

Primary LanguagePython

Description

Computer vision project for YOLOV8 implementation.

Objective

Recognize and distinguish the Durham College logo from the environment in each test image.

Steps

  • Open terminal or other command-line interface
  • Create and activate a virtual environment (venv)
  • Clone YOLOV8 (ultralytics) to working directory
  • Initialize a YOLO object with pre-trained weights file yolov8n.pt
  • Start the model training by specifying required parameters such as number of epochs (300) and patience (50)
  • Save model to train directory
  • Run test on the validation set and adjust previous values accordingly, for performance
  • Run test on the testing set

Results

The model accurately recognized the Durham College logo in most of the images presented in the testing set.

Note

Only test photos (predict) are uploaded due to file size restrictions on GitHub.