oki-aryawan / flower-detection-yolov5 Goto Github PK

This is my first deep learning project using YOLOv5. In this project I made deep learning model to detect 2 kind of flower in my house yard. There are 2 types of flowers (hortesia and mavavicus). In total I used 104 images that I took and labelled by my self.

• PyTorch
• YOLOv5
• Google Colab
• RoboFlow
• LabelImg

• Donwload my anotated images
• Create dataset using Roboflow
• Complete tutorial How to Train YOLO v5 on a Custom Dataset

• Clone the YOLOv5 repository by Ultralitics

git clone [email protected]:ultralytics/yolov5.git

• Open the yolov5 directory using IDE or code editor and setup the Python environment (I used Python 3.8 and Pycharm Professional Edition as my Python IDE)
• Get ready with all of the Python librabry that required

pip install -r requirements.txt

• Download my weights and yaml file from dataset directory
• Put those files in the yolov5 directory in your local computer
• Run the detect.py file using my custom weights

python detect.py --weights flower_07042022.pt --source 0  # webcam
                                              img.jpg  # image
                                              vid.mp4  # video
                                              path/  # directory
                                              path/*.jpg  # glob

• Result file will saved at runs/detect in yolov5 directory