Skip to content

Latest commit

 

History

History
32 lines (21 loc) · 1.25 KB

README.md

File metadata and controls

32 lines (21 loc) · 1.25 KB
Raw

YOLO-Custom-object-detection

In this project I used YOLOv5 model to train and predict custom objects such as malt seeds. Image dataset was splitted into two classes (acceptable and not acceptable)

  • Link to YOLOv5 official website YOLOv5

Used libraries

  • Python
  • OpenCV
  • All libraries included in the YOLOv5 documentation

Project stages

  1. At this stage it was necessary to collect the right images. The images come from personal database.
  2. Labeling objects in images. In this I used LabelImg
  3. Extracting data from XML files ---> Jupyter Notebook
  4. Conver labels information into DataFrame format ---> Jupyter Notebook
  5. Split data into train and test sets ---> Jupyter Notebook
  6. Create YAML file ---> Jupyter Notebook
  7. Training model ---> Google Colaboratory
  8. Get predictions from YOLO model ---> Jupyter Notebook

Examples

malt_detector

output2